Mechanization

By mechanization we mean the wave of mediatization through which the media environment became a mechanical one. In Europe and Northern America, the printing press – invented in its mechanical form around 1450 by Johannes Gutenberg – can be understood as the main origin of this process. However, we have to bear in mind that this invention is not just a European fact: already between 700 and 750 block printing arose in China Figure 1 and Korea, and the first experiments with ceramic letters were undertaken around 1040 in China. Later these techniques were developed further in Korea using bronze letters ().

The printing press did not change the appearance and structure of the book, but rather ‘mechanized’, indeed materialized through technology, a previous shift in how text was organized into a new form of production (). In this sense, the invention of the printing press was less an event of a single ‘communication revolution’ () and more a ‘long revolution’ (), or, maybe better, an important step in a long-term process of mechanization. This mechanization involved an increase in the co-existence and co-influence of mechanically printed media with handwritten manuscripts and other media (), but it also made a number of ‘new’ media possible: besides the printed book these were pamphlets and broadsheets, later the newspaper (). Interestingly, these media – mechanical or not – influenced each other in various ways. For example, printed treatises on the art of letter-writing stimulated the handwritten letters in early modern Europe. And for a long time manuscripts remained in use as they offered the chance to communicate controversial issues to a wider audience, which was embedded in practices of rewriting and ‘publish-ityourself’ ().

It would be too crude therefore to equate the invention of the printing press with a leap into a media environment consisting only of books and newspapers. The media environment remained diverse and was characterized by interactions between various media (). However, print became fundamental for the media environment because it changed the interrelation of other media, a change that has to be understood in a wider context. This mechanization of the media was a specific variant of other forms of mechanization like, for example, the invention of the mechanical clock, the railway, and machines in factories – just to mention some important further innovations that resulted in the ‘machine speed’ of the nineteenth century (). Therefore, we have to understand mechanization within the overall contradictory process of industrialization ().

To understand what such a fundamental qualitative change of the media environment meant for the everyday, we have to look at audiences: that is, people and their media use. If we consult historical studies, in Europe already within the seventeenth century the readership of newspapers (and books) was quite diverse, including the aristocracy, scholars and state officials, but also merchants, craftsmen, soldiers and women (). While the absolute number of copies of a newspaper in these times was a lot lower than today (between 300 and 400 copies), each copy had between ten and thirty readers (). Media appropriation was much more a collective venture than an individual one. An expression of this is the foundation of various reading clubs to share the costs of subscriptions and to have access to a wider variety of newspapers, journals and books. Starting around the seventeenth century, women became more and more part of the readership, and various journals and books were dedicated especially to them, building up their own communities of readers ().

While it is obvious that mechanization changed media environments substantially, we must be very cautious in attempting to deduce from this that the communicative construction of the social world was transformed in a homogeneous way. Rather, we can say that, with mechanization, media environments became more diverse and complex. They involved an increasing number of different media based on the mechanical reproduction of print: mainly broadsheets, pamphlets, books, posters, newspapers and journals. These print media were interrelated with non-print media. Mechanical reproduction offered through its standardized reproduction the possibility of reaching a wider group of people via one kind of media outlet. As a result, communication on a larger scale could ‘disembed’ () from the here and now. Hand in hand with this, new practices of communicative construction emerged, and it is a matter of contextual analysis to decipher these practices and to investigate under which historical circumstances and in which contexts they are located.

Within this plurality of possible transformations, mechanization helped to ‘thicken’ more extensive communicative spaces by intensifying trans-local communications. In Europe, these spaces were related historically to the rise of (national) states, something that is reflected by the concept of the (mediated) ‘public sphere’ (). The thickening of these national communicative spaces supported the emergence of ‘modern societies’ () with their national ‘imagined community’ (). But it is important to remember that the late Benedict Anderson’s model of the imagined community is just as much an account of the rise of new nations under or against colonialism (Thailand, the Philippines), and that the nation is only one type of imagined community made possible by print communication (). The thickening of translocal communications can also refer to completely different social entities: dispersed scholarly groups of experts, religious and ethnical diasporas, or other groups with distinctive world-views, can just as much be associated with the articulation of social conflict as the stabilization of nation-states. In Africa for example the emergence of the printed press was closely related to colonialism, that is, the attempt to impose a new colonial state on pre-existing societies (). This was a long-term process and involved, on the one hand, the construction of a ‘united “imagined community” of white settlers’ as a citizen public and, on the other hand, the encouragement of a ‘black readership aimed at entertainment’ () – with all its related problematic constructions. It is here that we start to see the consequences of a transformed media environment for larger possibilities of social change, that is, for media’s role in the construction of social reality itself (mediatization).

Electrification

A second wave of mediatization is electrification, which again changed the whole media environment, and also transformed mechanical print to a higher level of reproduction. In contrast to mechanization, electrification cannot be related to any single important media technological innovation like print, but to multiple ones (see figure 1), most importantly the telegraph, telephone, gramophone, turntable and audiotape, film, radio and television. By the term ‘electrification’, we mean the transformation of communications media into technologies and infrastructures based on electronic transmission. In electrification, media of various sorts became increasingly embedded in wider technological networks: in electricity grids, cable networks, networks of directional radio, and so on. This move is also a further step of interrelating media closer with each other, and so increasing the interdependences within the media environment.

The electric telegraph was invented in the 1830s. After the 1850s an intense wiring of the world began, starting from London and Paris, and having a first peak with the successful transatlantic submarine cable in 1866 (). However, this infrastructure of electric communication was far from egalitarian, centred first in the UK and later the USA, and in the hands of a limited number of companies funded by capital from particular investor countries (). The basic technologies of the telephone go back to the nineteenth century, and already in the 1920s long-distance calls became possible. The development of wireless telephone technology – radio telephony – began at the end of the nineteenth century (), with the first mobile train phones in Europe in 1926 and first mobile car phones in North America in 1946.

If we consider electric telegraphy and telephony, the striking fact is that electrification was in the beginning predominantly related to what is called ‘personal communication’ and not to ‘mass communication’ (). Initially, the benefits went to selected elite groups, especially in the fields of the military, economy and government (). In addition, the electric telegraph transformed print journalism as it made possible quicker access to transnational and transcultural information, which became first important for war coverage ().

Electrification involved incremental moves from mechanization, through partly electrified media, to fully electronic media. This becomes apparent if we refer to audio communication, where we have a sequence of innovations like the gramophone, turntable and audiotape. While the phonograph and gramophone were mechanical, the turntable was an electronic audio device for producing sound from the combination of phonograph and record. Audiotapes followed in the second half of the 1920s, as commercial products in the mid 1930s, again with many mechanical parts. In visual communication, we once again notice electrification on the basis of a sophisticated mechanization. Photography in the 1830s was a mechanical-chemical procedure; its electrification started in the 1950s through the use of electric parts in cameras. In a much faster way, film production also moved into electrification. Being in the beginning a mechanical-chemical medium like photography, the projection of films in cinemas made strong lights necessary as well as electric motors to drive the film reels. A further step into electrification was sound films. Over time, we have come to consider film as an electronic medium.

Typically, however, electrification is thought of in terms of broadcast media, that is, radio and television. Developed as technologies at the end of the nineteenth century, the first radio broadcast took place in 1906. However, not before the end of the First World War and the beginning of the 1920s did radio become a more widespread medium. In the beginning the medium was technically open for a mutual transmission of private messages; this fascinated Bertolt Brecht, for instance, who was interested in its capacity as a horizontal ‘communication system’ (). Driven, however, by commercial and political interests it quickly became sender-centred, being legally regulated and technically materialized as broadcasting (). Already by the 1920s and 1930s, radio was established in Europe, North and South America, and also in Asian and later African countries (). Television goes back to a late nineteenth-century vision of making the transmission of (moving) pictures by telegraph possible (), but regular television broadcasting did not start until the mid 1930s in Germany, the UK and the USA (). The main breakthrough of television happened in the late 1940s in the USA, and in the 1950s and 1960s in Europe and many other parts of the world. Television changed its character continuously, becoming in the 1970s a coloured medium with increasingly new and original formats, additional related devices (remote control, VCR), and the spread of satellite television across borders in the 1970s and 1980s ().

These historical moments demonstrate again how by electrification the whole media environment became transformed. ‘New’ media emerged, but also ‘older’ media like the press underwent change, too. Electrification was the trigger for a deep qualitative change of the whole media environment within a rapid sequence of media technological innovations, perhaps having its first peak at the beginning of the twentieth century – a time in which contemporaries already imagined a ‘wireless age’ (). Across the world, electrification was an enormous endeavour, with high public and private investments being necessary to build up the appropriate technical infrastructure, first by landline and cross-Atlantic telegraphy cables, later by broadcasting stations and cable networks.

Electrification can nonetheless be seen – at least in its beginning – as an extension of mechanization, but with the key difference that it involved a deeper interlinking of media through a new technological infrastructure: the electricity grid, cable and broadcasting networks etc. Single media came to appear less ‘independent’; increasingly, technical capacity was not only on the producers’ side of the media, but also on the users’ side. Therefore, at a fundamental level electrification can be understood as a move into a deeper, and technically more intertwined, media environment.

How should we understand the social and cultural consequences of electrification? We must again be very careful not to oversimplify the related social and cultural transformations around the globe. It is too crude to run together all forms of technical connectivity by assuming they had a single global consequence (). It would be too reductionist also to assume that electrification of media caused something like a ‘global neighbourhood’ (). However, proceeding in a more open way, we can notice at least three interdependencies involved in electrification, as it came to shape the media environment.

First, with reference to all kinds of produced media communication, electrification made possible simultaneous transmission of media content across space. Crucial here was the emergence of powerful media organizations in the film, but especially broadcasting, industries. The connectivity they enabled through their production cycles generated shared rhythms of simultaneous experience and new narratives of commonality, whose clearest form was the ‘media event’ in which there emerged the ‘rare realization of the full potential of electronic media technology’ ().

Second, electrical transmission made new forms of near-instantaneous, reciprocal communication feasible across long distances: the new possibility of simultaneity in personal communication across space. Various media like telegraph, telephone etc., some of which were widespread before the electronic mass media, support this. As a consequence, this wave of mediatization is not only related to the emergence of new types of media organizations, but in addition and much more far reaching it made possible completely new kinds of social institutions. For example, already with telegraphy in the 1860s, a globalized economy began to emerge, when ‘the first tickers clicked out stock prices’ ().

Third, these forms of communication offered possibilities for constructing cultures in new ways across space and time. Electrification brought media that supported a further thickening of translocal spaces of communication. It was a wider support for stabilizing national communicative spaces and at the same time for a possible social and cultural differentiation within and across those spaces then, for example among diasporas: that is, transnationally dispersed cultural groups which maintain their relations in real time through electronic media (). But this is only one example of a broader possibility: the establishment, through structures of simultaneity and shared rhythms of media flow, of translocally configured cultures of many sorts, including at the most general level the emergence of large-scale popular culture since the 1950s.

We are not, of course, saying that electrified media environments were, or are, the same across the world and would have resulted in one kind of global change at the social and cultural level. Other dynamics took place in Nigeria, for example, where videocassettes and the infrastructure of privacy had an important role for building up counter-spaces and alternative collectivities (). In general, ‘indigenous media’ () supported a range of possible dynamics and consequences: coming together through media as a collectivity, constructing shared memories through media, and gaining a ‘voice’ in public discourse. In Latin America, there has been a ‘hybrid history’ () of appropriating technologically based media of communication: beyond a linear ‘Western’ modernization this has supported processes of transculturation, especially in urban areas. These media environments made possible not just nationalized but also highly de-centred and globalized processes of communicative construction, each of which must be analysed contextually.

Digitalization

Digitalization is the third wave of mediatization within the last few centuries, and is typically related to the computer, internet and mobile phone.⁶ All three are key inventions for this wave of mediatization, but again we should be reserved about describing them as ‘revolutions’ (): like all other inventions, they are the result of complex distributed social processes of making. One perspective on this history sees algorithms and software as the fundamental level of digital media;⁷ another emphasizes the perspective of politics in the history of digitalization;⁸ while a third brings out the cultural contexts and ‘pioneering’ groups behind these developments.⁹ However, we want to start our analysis by moving the development of ‘the internet’ itself into the foreground. As we emphasized earlier, a main characteristic of waves of mediatization is the intensified interrelatedness between media that they involve: the internet is the infrastructure that makes possible the linkages between contemporary media devices with mainframe computers, large data centres, and – in the near future – social robots and autonomous systems like self-driving cars, as well as links all our activities on countless digital platforms.

The history of ‘the internet’ has been told many times. Everyone knows that it emerged from the research arm of the USA’s military establishment through its connections with university research labs. In this sense, it is a strong example of how developments for which ‘the market’ claims ultimate credit usually derive from deep underlying subsidies by the state and other forms of public institution (). Particularly important is the combination of steps – some state-led, some driven by markets – as a result of which in 2015 a small number of corporations that loosely can still be called ‘media’ – Google, Facebook, Apple, Instagram, perhaps also Twitter and, in China, Alibaba – can through their ‘platforms’ act directly on the world of consumption and the world of everyday social interaction. The stages involved in that development are worth setting out more fully.¹⁰

First, there was the building of ‘distributed’ networks of communications between (initially very few) computers through the innovative process of ‘packet switching’, as a means, initially, to ensure more secure forms of communication when facing military attack (). The next stage was the development () of the TCP/IP protocol¹¹ for connecting up groups of already linked computers into a wider network, originally implemented in the early 1980s, and leading by 1989 to an ‘internet’ of around 160,000 computers in the public sector. The defining stage was the emergence of the World Wide Web, which occurred through two key inventions: the idea that texts could be linked together if they were associated with ordered sets of ‘metadata’ called ‘hypertext’, and CERN’s Tim Berners-Lee’s formalization of the means to ensure the reliable transmission of hypertext between linked computers (HTML, or Hypertext Markup Language; HTTP, or the Hypertext Transfer Protocol; and URL, or Universal Resource Locator, which provided an address code for each hypertext file). Taken together, they enabled the proposal in 1990 for a ‘web’ of files on networked computers and the first system for ‘browsing’ the domain of those texts (the World Wide Web), and the first ‘web’ site in November 1991 (info.cern.ch). This publicly subsidized development had produced by the early 1990s the skeleton of a connective infrastructure, but this was not yet linked to everyday commercial activity, or even non-specialist everyday use.

A rather different and accelerated sequence generated the deeply commercialized internet and WWW that we know in the second decade of the twenty-first century. In 1991, NSFNET was closed and the internet’s operations handed over by the US government to commercial providers. From this new starting-point, the first commercial web browser (MOSAIC) was developed by a researcher, Marc Andreesen, who had left the public research sector (the NCSA, or US National Centre for Supercomputing Applications) to found his own commercial company called Netscape Communications, which went on to produce and mass-market Netscape. In the late 1990s and early 2000s, the means to access the exponentially growing domain of internet-linked files shifted from the model of managed directories (Yahoo) to Google’s algorithmically based model of indexing pages based on a hierarchy developed from, at its most basic, counting the number of links in to each internet page. The distinctiveness of Google was that, rather than process a bounded and finite directory, its operations were, in a technical sense, recursive, with each new link increasing the data over which its calculations ranged, and increasing the mechanism’s power, without end. This key innovation facilitated internet use considerably, and emerged alongside the diffusion of small desktop computers and then laptops as means for accessing the internet easily.

Building on the huge success of its Google search engine, Google bolted onto it the foundations of a much more robust commercial infrastructure: a new model for advertising tied to terms searched through the Google search engine (‘Google Adwords’) and a system of live-auction advertising (Adsense), which together opened up a new basis for the marketization of online ‘space’.

The crucial next step was the independent development of ‘smart’ mobile phones with the capacity not just to provide access to the traditional modalities of phone use (talking/listening, plus the key discovery of the mobile phone device: the sending and receiving of text, or SMS), but also to the domain of the World Wide Web. The emergence of ‘smart’ phones was followed relatively quickly by the design (led by Apple, but quickly followed by most other smartphone providers) of applications, or ‘apps’, installable on each phone, to provide accelerated and simplified access to particular domains of web data. A further crucial step involved the emergence () of a new type of website which provided platforms for hundreds of millions of users to network with each other, but within the parameters of form and content management designed by that platform’s owners: so-called ‘social media networks’. This is the twenty-first-century replaying of the emergence of possibilities for lateral communications which Craig Calhoun () noticed in the history of nineteenth-century communications, but this time harnessed to the capacity for every message, in principle, to be directed at, or subsequently re-circulated to, the whole domain of the WWW through acts of ‘mass self-communication’ ().

The result of these cumulative and interlocking steps is a strikingly complete transformation of ‘the internet’ from a closed, publicly funded and publicly oriented network for specialist communication into a deeply commercialized, increasingly banal space for the conduct of social life itself. The sheer size of the data transmissions now occurring via the internet has generated entirely unprecedented infrastructural demands (especially for storage, but also to support data processing) that are being satisfied, once more, not by public interests, but by a small number of private corporations that dominate the so-called ‘cloud’ ().

Again, the wave of digitalization was not only a matter of so-called ‘new’ media. ‘Old’ print and electronic media also became more and more digital. This is very well documented for print, especially the book and newspaper industry (). Indeed, television and film also became digital, from the perspectives of both production and use, most recently with internet television, second screens and digital film projectors. As digital products they could be pirated and reached further audiences and users across the globe who build ‘underground networks driving globalization from below’ (). Related to digitalization is also a fundamental shift in the business models of media derived, in part, from shifts in how advertisers can reach their target audiences, that is, increasingly through personalized tracking of individuals and related data collection (). Another emerging trend is the increasing embedding of digital media capacities in objects that accompany everyday practices: ‘virtualized media communication’, such as computer and mobile phone ‘assistants’ simulating a living communication partner, or ‘social robots’ as ‘artificial companions’ in our lives (), something that more and more becomes realized through software ().

How should we understand the cultural and social consequences of these changes? It is certainly not enough to talk, as we often do, of media saturation, of even supersaturation (). The saturated nature of the media environment was already being noted back in 2003, that is, towards the beginning of the wave of digitalization, and long before its peak (). Some writers have sought to accentuate the abruptness of the digital transition, arguing that the new capacities of receivers of communication to send communications from the same device, and often as part of the same communication cycle, have led to the disappearance of the ‘person formerly known as the audience’ (), or the emergence of the hybrid ‘produser’ (). Undoubtedly, the starting possibilities for our relations to media have changed, but it is unhelpful to read the resulting forms and patterns of use in too polarized a way. Audiences are able to do an extended range of things with media (): for example, their commentary on media is now available to be folded back into production cycles in ways that were previously impossible. But media produced by media organizations have not disappeared, and today’s forms of media saturation are only an intensification (driven by the capacities of mobile phones and social media networks) of – not a fundamental break from – cultural forms that emerged late in the wave of electrification with the late 1990s growth of reality media and celebrity culture in many parts of the world (from Brazil to Korea, from Lebanon to South Africa).

We would do better to look for change at the level of the increasingly individualized patterns through which people can access, follow and comment upon (in other words, ‘actualize’) what we propose to call the media manifold (defined shortly). As yet, we do not know much about those figurations, except that they are much more varied than were possible in the electronic wave of mediatization, when most media content came from a limited number of synchronized central sources, when opportunities for media production were very limited and tied to the operations and gatekeeping power of those central resources, and when commentary on media was almost always lost to the air.

Put more generally, digitalization involves a further deepening, both in the connectedness of the infrastructures on which media-related practice depends – so, for example, the digital is now dependent on the development of Wi-Fi and other mobile services – and in the layering of connected media practices in which individuals or groups are now routinely engaged. Media environments have become characterized increasingly by ‘convergence’ (),¹² meaning less a merger of all media devices into one kind of super-device, and more convergence at the level of the ‘data’ or content which, being digital, becomes communicable across multiple devices, some new, some older.

Living in the midst of this third wave of mediatization, it is obvious that it has far-reaching consequences for the communicative construction of the social world. But it must be understood merely as the provisional climax of this latest wave of deep mediatization. We now see signs of a further wave of mediatization, a wave that is related to data. If we consider how far datafication changes the way we produce knowledge (see Chapter 7), how deeply it is related to the constitution of the self, of collectivities and organizations (see Chapters 8, 9, 10), these media-related changes may well be more far-reaching than those we currently associate with digitalization. Certainly, we can expect a further deepening of the relations of interdependence between media and between people, when an increasing proportion of communication relies on infrastructures of communication based on the collection and processing of data. The degree to which this is the harbinger of a further qualitative shift of the whole media environment must remain, in part, an open question, but one to which we will return at various points in the chapters that follow.

3.3 Deep Mediatization and the Media Manifold

Mediatization came in waves – mechanization, electrification, digitalization – which each changed the whole media environment fundamentally. But since it is the whole media environment we are talking about, such mediatization waves cannot be understood as the ‘diffusion’ () of one dominant medium, and it is much too crude to say that we have always lived in a newspaper, television or internet/mobile phone era. To understand mediatization, we must understand it as a process of the increasing deepening of technology-based interdependence. This deepening has two senses; first, that over the past 600 years an acceleration of technological innovations in media has taken place; and second that, over the same period, media have become increasingly relevant to articulating the kind of cultures and societies we live in, because of media’s changing role in the conditions of human interdependence.

Acceleration of change means in a basic sense that the sequence of more or less fundamental technological innovations in the field of communications becomes shorter. While for example the interval between the invention of the printing press and the printed newspaper was around 150 years, telephone, film, radio and television were invented within fifty years, and the innovation of various digital media took place within thirty years. In a more complex way, comparing the three waves of mediatization – mechanization, electrification and digitalization – the change of the whole media environment in which they resulted took place in increasingly short time-periods, linked to the increasing causal weight of pre-existing media in shaping the new interconnected media environment that emerged. Both can be related to a more general phenomenon: what Hartmut Rosa has called the ‘progressive acceleration of social change’ (). If we recall innovations in media technologies presented in the earlier figure, we can follow Rosa’s proposal of a transformation ‘from an intergenerational speed of change in early modernity through a phase of approximate synchronization with the sequence of generations in “classical modernity” to a tendency toward an intragenerational tempo in late modernity’ (). But once again, we should avoid tying this acceleration automatically to ‘European’ modernization: with that caveat, Rosa’s metaphor may help us to grasp the acceleration of media’s role in processes of globalization themselves.

Such cumulated waves of mediatization have resulted in a media environment that is unique in its present form: many communications media – even the stone tablet and manuscript – have not disappeared but kept a special functional role, partly within the arts. Some early electronic media, like vinyl records for example, even have a revival (). Beside that, a further landscape of digital media has been established – the mobile phone, online platforms and computer games are just some examples – while older electronic media such as the television, radio or cinema themselves became digital; all of this in a sustained and ever more distributed infrastructure. There remains, however, a significant challenge in grasping the complexity of the contemporary media environment adequately.

Underlying these changes has been a deeper change, that media’s degree of technical interrelatedness has considerably increased from mechanization to electrification, and then again from electrification to digitalization. The technical interrelatedness of mechanization was limited insofar as technology was mainly used on the side of media production and far less on the side of media use. Think of the printing press: the distribution of printed books, magazines and broadsheets took place through physical mobility, and the use of these print media did not involve any use technology at all. This changed with electronic media: media became technologically intertwined, with the wider power grid of electricity distribution being a precondition for electronic media’s subsequent distribution, and media themselves becoming dependent on their own technological infrastructure, including broadcasting networks, cable networks, radio networks etc. This technological relatedness partly became manifested in new kinds of devices that integrated various ‘different’ media into one terminal, like for example ‘compact systems’ combining radio, audio cassette, records and sometimes even television into one end-device at the user’s side. In the broadest sense, digitalization has been a further move in the direction of technological interrelatedness:¹³ through digitalization, it became possible to move various kinds of ‘content’ across a connective infrastructure, the internet. In turn, our devices for accessing the internet may no longer be specialist computers, but can be multi-purpose devices such as smartphones and tablets. Even so, while in such a media environment most devices are based on aspects of computer technology, we do not yet have a convergence into one ‘meta-device’.¹⁴ What we have instead is a deep technical interrelatedness of the increasing variety of different devices; this is what makes connectivity so omnipresent as a requirement of our present times. In addition to this, the character of each medium is more and more defined by the particular software and calculative functions on which its underlying functions are based, and not just by the technological device as such. What we call a ‘mobile phone’ is a device that can ‘represent’ to us multiple media, and can even be extended by adding further apps that give filtered access to other media streams.

The resulting increased interrelatedness of contemporary digital media cannot just be considered as a question of choice: which medium an individual selects for what purpose. The deepened interrelatedness associated with the wave of digitalization defines a new kind of media environment different from earlier media environments. We need therefore to develop the right analytic tools to grasp what is distinctive about this environment, and our relations to it.

As the wave of digitalization has proceeded, various kinds of description have been proposed to capture the new types of interrelation that have been emerging. Some terms put emphasis on the interchanges of content between various media; for example, ‘remediation’ (the ‘representation of one medium in another’ (), ‘transmediality’ (), and ‘spreadable media’ of ‘viral’ communications across various platforms (). Other concepts foreground the appropriations of multiple media by users: the terms ‘cross-media’ (), ‘media repertoires’ () and ‘polymedia’ () are examples of this. Such a terminology understands the contemporary media landscape as a ‘composite environment within which each medium [and its use] is defined relationally to all other media’ (): ‘media repertoires’ are not just a sum of the media a person uses but the meaningful relation between them in everyday practice. Deep mediatization on this view is marked by user practices that move across a variety of media.

Such concepts hint at the fundamental transformation of the social world that flows from deep mediatization and they are correct in trying to grasp the relations of interdependence that characterize the digital media environment. However, they do not capture the interrelated complexity that is characteristic of the digitalized media environment as a whole. To capture this, we offer the concept of the media manifold.¹⁵

The term ‘manifold’ comes from mathematics, specifically topology, where it refers to a topological space in many dimensions that can be adequately described by a shape in a lesser-dimensional (for example, Euclidian) space. So the earth is a three-dimensional shape that can, with reasonable fidelity, be reduced to a set of two-dimensional ‘maps’ of parts of its surface. Deleuze used this notion to emphasize the open-ended complexity of the world, but his emphasis was rather on how that order escapes any simple reduction to a model (). The Deleuzian usage however seems to lose touch with the two-level aspect of the manifold concept that, we would argue, is most useful in grasping how we stand now with media: that is, the relation between a many-dimensional object and the approximation to that object in another object with fewer dimensions.

Our suggestion is that this double concept (the ‘manifold’) well captures the doubleness of our embedding in today’s extremely complex media environment. The set of media and information possibilities on which a typical social actor, at least in rich countries, can now draw is almost infinite, and organized on very many dimensions. But it is in fact a reduced set of possibilities from which we choose every day: that reduced set is how, in practice, we actualize, for daily usage, that many-dimensional media universe.¹⁶ What we then do with media on particular occasions actualizes, in turn, that reduced selection, and comprises itself a distinct and important level of variation. There are therefore three levels in operation; but here, as we try to understand our relations as social actors to the wider universe of media, it is only the first two levels (and their interrelations) that most concern us. The relation between this reduced set of daily options and the infinity of options in principle available is what we mean when we talk of our relations with a ‘media manifold’, holding onto the idea of ‘manifold’ as a many-dimensional object that can be captured adequately in a lesser number of dimensions.

The term ‘media manifold’ enables us to keep in view both the social actor’s position within a much larger institutionalized environment of interdependent media and the situated complexity of that actor’s everyday choices of media. We need to understand both, and their interrelations, since the dynamics of that wider environment, particularly its overriding pressures towards datafication, are of major consequences for all actors and for the organization of social life as a whole.¹⁷

In the next chapter we turn to how we can think sociologically about the consequences of our relations with today’s very complex media environment for the construction of the social world.

Understanding Figurations

At its most basic, figuration is ‘a simple conceptual tool’ () to allow a thinking in which ‘the individual’ and ‘society’ are not treated as antagonistic. A figuration is a kind of ‘mode[l] of processes of interweaving’ (), a more or less stable interaction of individuals which produce in this interrelation a certain kind of social meaning. One can take a football match or card game as examples to explain what a figuration is: the people involved form a figuration as their interactions are oriented to each other in an interdependent way. The game is the ‘outcome’ of the interrelated practices of the involved individuals and their ongoing process of playing. It is ‘more’ than just the gatherings of individuals but at the same time not something ‘beyond’ them. Or put differently: figuration means the ‘changing pattern created by the players as a whole – not only by their intellects but by their whole selves, the totality of their dealings in their relationships with each other’ (). If we follow Elias, the fundamental idea is to understand more or less durable social formations of humans as figurations: they are constituted by the interdependencies and interactions of the involved individuals and can be characterized by a certain ‘balance of power’ (), that is, power relations. The boundaries of each figuration are defined by the shared meaning that the individuals involved produce through their interrelated social practices, which is also the basis of their mutual orientation to each other.

The idea of figuration has parallels to the concepts of network. Elias himself emphasized repeatedly the relation of his idea to the structural category of the network, for example when he describes the interrelation of individuals in figurations as ‘networks of individuals’ (). Elsewhere he argues that ‘social figurations’ are a kind of ‘human network’ () or ‘ordered network’ (). His distinctive take on networks becomes clearer when we consider his detailed analysis of different figurations. In these analyses ‘network’ is an important term to describe the relations of the intertwined actors, especially when it comes to what he calls ‘game models’ of figurations (). But Elias goes far beyond network as a metaphor. The latter is helpful to analyse the interrelations of actors in a figuration and to describe some fundamental characteristics of them as ‘models of interweaving’ (). But the network does not yet describe the co-oriented practices of sense-making in a figuration. Therefore, a figurational approach means much more than describing the network of actors: it means considering also the power relations, the characteristic roles in a figuration’s actor constellation and the overall meanings that are thereby produced.

Figuration has also certain parallels to assemblage, as both describe the social world in terms of processes of individuals’ mutual interweaving. Indeed Latour explicitly refers to the idea of figuration (), although without discussing Elias. However, figuration for Latour is not only a figuration of human actors as ‘there exist many more figures than anthropomorphic ones’ () and we must, he argues, analyse figurations of humans and non-humans (). On this view, methodologically grounded research into assemblages might itself appear to be nothing more than the investigation of certain figurations, but with the crucial difference that Elias, when reflecting on the role of objects and technologies in figurations (), always makes a clear analytical distinction between objects and human actors. This is an important point to which we will return.⁶

A figurational approach allows us to integrate the strengths of network and assemblage analyses – their focus respectively on constellations of actors and socio-materiality – while going further in explaining how the embedded complexity of communicative practices works. A figurational approach indeed has all the strengths of complexity theory, especially its sensitivity to nonlinear causal processes, respect for the contingency of process and the possibility of multiple outcomes, and insistence on the importance of emerging relations rather than fixed objects. But it manages to achieve this directly within the language of sociology and social theory. Indeed Elias’ notion of figuration was introduced as a way of changing the social sciences’ ‘means of speaking and thinking’. Sociology’s inherited vocabulary, Elias argued, freezes processes into things, specifically things such as ‘norms and values’, ‘structure and function’, ‘society’ and ‘individual’ (). Instead, we need to grasp ‘the special kind of order associated with processes of social interweaving’, which means ‘start[ing] [. . .] from the connections, the relationships, and work[ing] [. . .] out from there to the elements involved in them’ (). A ‘figuration’ is ‘a flexible lattice-work of tensions’ () which, while its development is open-ended, remains regular and interconnected enough to form something relatively stable, and so worth analysing as a pattern.

The force of the figuration lies in its basis in an understanding of relationships of meaning. Elias wrote that ‘the behaviour of many separate people intermeshes to form interwoven structures’ (). The term ‘intermeshing’ sounds like mere metaphor, but the metaphor precisely captures a number of things: first, a feedback loop (so much is common with complexity theory, systems theory and assemblage theory); second, a feedback loop whose paths are comprised of interlocking practices, acting back on themselves; third, practices that interlock because, as meanings, they are in a mutual relationship, answering, inviting, challenging, questioning and so on. The elements of a figuration only have a common form (a con-figuration), because there is something at stake in them, something that matters (is meaningful) to the actors involved.

We can see plenty of examples of figurations in times of a deep mediatization. Many are new, involving us in new types of action in which new things are at stake. Figurations may involve, at one end of the spectrum, chains of photo-exchange on Flickr and information or debate threads on Twitter and (at the more elaborate end of the spectrum) the whole interlinked ecology of platform-based message circulation in which, for example, celebrity promotions, or friendship building, or project promotion now evolve. We cannot grasp the dynamics of these new processes unless we understand them as more than associations of heterogeneous elements (assemblages) and more than structures of linkages (networks). The figurations of online practice comprise an open (expanding) set of spaces for interaction and dependency, in which we are enmeshed, as we try to go on doing what we ordinarily do. Their dynamics also cannot be understood except within a larger strategy to build an infrastructure for sociality online that is a feature of the wave of digitalization.

The Basic Features of Figurations

The basic idea of figuration already offers many of the conceptual tools we need for analysing the complexity of our life with media (). As a figuration coalesces, it begins to stabilize relations between what until then were disparate sites of practice. However, Elias could not foresee the deepening of mediatization – with its waves of mechanization, electrification, digitalization and now also (possibly) datafication. As a result, for this context of deep media-based interdependency, we need to specify three distinct dimensions of how figurations stabilize: their relevance-frames, actor-constellations and communicative practices, each of which is founded, in part, on relations of meaning. Let us explain. First, each figuration has certain relevance-frames. By this we mean that the people involved in a figuration have a common orientation to a shared ‘purpose’, whether it be as a family, a group of friends, a collectivity or as users of a particular digital platform. The relevance-frames of a figuration express its social meaning as a distinct way of acting together. Sharing this set of relevance does not of course rule out conflicts or disagreements. There can be, for example, many conflicts in families as in any other kind of figuration, but the point is that these are understood as family conflicts. When it comes to our living with the media manifold, new questions arise: do older relevance-frames get transformed by our relations with the contemporary media manifold? How far are certain relevance-frames related to the new media environment? And does this enable new types of figuration to arise?

Second, each figuration involves a distinctive constellation of actors. This phrase has a double meaning. First, the individuals in a figuration are not just a random accumulation of individuals. They are related to each other in typical ways, for example, because they have specific roles in the figuration (parents and children in a family). There are certain social definitions of that relatedness (close or less close relationships between friends etc.). Second, a figuration is a constellation of ‘human beings’ (). This does not mean that we should not consider objects and technologies – including media – as elements in figurations, but simply insists that objects and technologies are not part of the constellation of actors who understand themselves to be acting together in this way. Figurations can exist without objects and technologies, but they cannot exist at all without individuals. Therefore, its actor-constellation is more fundamental to a figuration than the objects and technologies involved in it. It is an arrangement of individual actions that is characteristic for that figuration, but the constellation is also open to change.⁷

Third, each figuration is based on certain practices that in turn depend on an ensemble of objects and technologies. Put another way, each figuration is based on certain distinctive practices of communication and a related media ensemble (). It is through the interrelated actions of such practices that individuals construct figurations: that is, figurations involve ways of doing certain things together, or in coordination, very often with and through media. The communications that arise around those practices contribute to the overall ‘meaning’ of the figuration. But we cannot understand the practices of such figurations without the objects and technologies we use in relation to them. While not being necessarily a constitutive feature of figurations – as relevance-frames, actor-constellation and practices (of communication) are – figurations typically come together with certain objects and technologies. Families, whether located at a single place or mobile (like migrant families), have certain possessions: household goods, maybe even apartments or houses. But they also have a variety of media to communicate with each other. Nowadays, this is even the case for homeless people for whom media can offer a certain ‘ontological security’ while being forced to live on the streets.⁸ The character of these properties has a lot to do with the specificity of these figurations and in addition supports a certain stability. A family with a great amount of property and a rich media ensemble can develop a much greater durability than families that primarily rely on personal relations. So it is not just technology that makes figurations more ‘durable’ (); it is physical objects of all sorts.

To sum up, the formation of figurations as patterns of communication in which something distinctive is at stake emerges through the interrelations between three dimensions: relevance-frames, constellations of actors, and communicative practices, that have, as their basis, a particular ensemble of objects and media technologies. These dimensions are relatively autonomous, but because each is involved in the situation in which action occurs, processes of acting together generally tend to reinforce them, and stabilize patterns of association between them. All these dimensions are based in relations of meaning. It is this attention to the distinctive consequences that flow from social worlds as communicative orders between human beings, built in part through regularities of communication and meaning that is missing from the notions of network and assemblage.

Power and Belonging

What of the larger social arrangements that emerge through figurations? We can approach this from two angles: power and belonging. In both cases, a double perspective is needed, asking on the one hand how figurations internally can be characterized by their power-relations and belongings, and on the other how figurations build power and belonging externally in the wider social world.

Each figuration has a distinctive actor-constellation, with certain characteristic power-relations. Elias wrote about ‘balance of power’ (), but that term can be misleading if we read it as evenness. Particular figurations are characterized by certain ‘conflict-ridden figurational dynamics’ (). But power-relations in figurations are not just a matter of positioning in the actor-constellation. They have a lot to do with practices of communication that make sense of these power-relations. Analysing power in figurations involves at least three levels: the positioning in the actor-constellation, the practices that support power-relations, and the inscription of power in the media ensemble. For example, as the analysis of gender relations in families has demonstrated, power within them correlates with the disposition of certain media technologies: the remote control, VCR etc. (). If we think about contemporary media technologies, the inscription of power with reference to gender even goes further: for example, the preselection of a certain gender (‘male’, ‘female’, …) when joining a specific platform predefines how as an individual one is represented and so can act through communication.⁹ And if we not only reflect on the internal power-relations of this figuration but also its external relation to other figurations, we have to consider how far the overall orientation of a figuration – its relevance-frames – is related to the power of this figuration more broadly. For example, being part of a certain collectivity – a group of male refugees or the community of a shared office – is related to the power that the members of this figuration have in the wider social world, to which media discourses about this collectivity.

Figurations are also closely involved in the construction of belonging (see the detailed analysis of collectivities in Chapter 9). Belonging can have a purely situational meaning. Think here about the figurations of games or the dance (Elias’ regular example): sharing orientation to common interactions creates a certain situational belonging of being part of, for example, the game or the dance. Think of the figuration of spectators in a stadium. In such a figuration we experience a deep feeling of communitization (Vergemeinschaftung: Weber ()), a situational feeling of being part of a crowd within an event (). However, this does not necessarily result in being part of a longer-lasting community (Gemeinschaft). In cases like these the deeply felt belonging is built up through the intensity of the figurations’ practices. Today, occasions for figurations with such an intense contextual meaning are more and more interwoven with media. The figuration of stadium spectatorship cannot be understood outside of its media ensemble: loudspeakers, scoreboard, display panels, and spectators’ parallel communications via their mobile phones and tablets. But even more media-related are figurations of music concerts or collective viewing of television and cinema in public places. And with digital media we have the new figurations of online gatherings – in chats, on platforms, through apps – through which we construct an intense situational belonging to each other.

Figurations and Media Infrastructure

At this point, we reach and need to go beyond the limits of the original concept of figurations as Elias developed it. Starting from the late 1960s to the 1980s – all before the emergence of the internet as an infrastructure for everyday social interaction – Elias offers basically a hydraulic metaphor for capturing the process whereby particular human beings become increasingly interdependent through the flow of meanings, promises, obligations and performances. The term captures well the length of these potential chains of interdependence, but tells us less about the sorts of entities that are bound into figurations. As a result, it tells us little – without at least the additional work we do later in this chapter – about the forms of dependence on system-infrastructure that are crucial to the very complex figurations characteristic of our relations to the media manifold. Elias only seems to conceptualize the figuration as an order that emerges from the moves of the individual themselves: ‘by figuration we mean the changing pattern created by the players as a whole [. . .] a flexible lattice-work of tensions [. . .] a fluctuating tensile equilibrium, a balance of power moving to and fro’ (). Yet, his own later reflections already by the 1980s acknowledged that media and communications technologies, and the large-scale systems they generate, were themselves already intensifying the complexity of the social, and its degrees of interdependence, generating ‘fluctuations in what might be called “social pressure”, in particular the “internal pressure” in a society’ ().

Certain types of figuration – associated with distinctive ensembles of media technologies – generate obligations and dependencies not just between individuals, but also between individuals and communication systems, obligations that are distinctive features of how we live within the media manifold, but which also characterize new types of figuration.

Let us start with the most basic examples. Take the figuration of a family: this figuration is now increasingly characterized by a distinctive media ensemble, whether it is particular patterns of watching television together or in sequence, particular uses of smartphones to organize family practices and interactions, the photographs we share in printed photo albums or (more usually) via online platforms. All these media contribute to the ongoing construction of a group of people as a family, with all its contradictions, conflicts etc. So too with the figurations of organizations where, with ‘ubiquitous computing’, both inside and outside its formal structure, it has become, some argue, ‘increasingly difficult to separate people’s interactions with other people from people’s interactions with technologies’ ().

But we have to go a step further: the media ensemble of a certain figuration is significant in the way it moulds its various practices of communication. Mobile phones and digital platforms mean a lot for families, when it becomes possible to locate family members wherever they are, when family members can be reached at all times, and when family members have reference to mediated representations of themselves as a family on a continuous basis. These material possibilities of contemporary media and related infrastructures not only offer better chances of sustaining the family figuration in a particular way; they also mould our practices of communication.

However, this is not a ‘one-way-street’ of media effects. Figurations produce certain needs of communication and therefore are an ongoing source for developing new media technologies and adapting present. The digital platforms we live with, for example, did not come out of nowhere. On the contrary, they responded to families’ needs to be reachable across space and time. ‘New’ technologies get appropriated within a figuration’s media ensemble (), quite possibly changing the figuration, and generating new demands for media, and so on in an endless feedback loop. This offers again chances for further media adaptations by ‘producers’ and ‘designers’.

Many of the most recent media developments refer to fundamental human needs of connection. This is reproduced, further institutionalized and materialized in the connectivity of digital platforms (). Understanding our lives with media means therefore grasping an ever more complex range of media-related figurations. But to do this fully, we need to conceptualize the transformed scale on which many contemporary figurations work.

Meaningful Arrangements of Figurations

There is however a second principle for scaling figurations: meaningful arrangements. Several times already we have emphasized that any attempt to understand the construction of the social world through communication is necessarily a theory about shared meaning production. We understand meaning at this point in the original sense of Max Weber () as Kulturbedeutung, that is, cultural meaning as actors in the social world produce it. In such a perspective the figurations of the social world also have – beyond any concrete links between actor-constellations – meaningful arrangements with each other. So, for example, national government agencies are figurations that are understood as ‘centres of power’ in relation to many other figurations. Such constructions of some figurations as ‘powerful’, ‘public’, and so on, and others as ‘weak’, ‘private’, and so on are only partly explicable through the interrelatedness of their actors or through the composition of ‘figurations of figurations’. For, in addition to such structural links, figurations in which we live also ‘hang together’ () with each other through relations of meaning, and this ‘hanging together’¹² derives from two distinct sources: certain discourses that connect these figurations and their meanings in the social world, and certain larger-scale relations of interdependency between domains of action (for example, the transport infrastructure and the economic infrastructure) that come to be associated with assumed relations of meaning. Both types of arrangement go beyond the simple inclusion of actors or figurations in a larger or more complex figuration (or figurations of figurations).

Peter Berger and Thomas Luckmann tried to explain something like such arrangements of meaning through the notion of ‘symbolic universes’ (). Symbolic universes were for them ‘meaningful totalities’ that share an overall understanding of the social world. They offer an ‘all-embracing frame of reference, which now constitutes a universe in the literal sense of the word, because all human experience can now be conceived of as taking place within in’ (). But this functionalist reading of how meaning contributes to order was never fully plausible even when they wrote, and it is absolutely implausible today in an age of proliferating meaning-making across countless national and transnational digital media.

Berger and Luckmann’s language of ‘symbolic universes’, if applied here, would suggest that all figurations fit neatly within functioning meaningful wholes. But this underestimates the variety of different discourses that relate figurations to each other by ‘telling’ or ‘explaining’ the ‘overall sense’ of certain figurations, and so, over time, construct larger-scale patterns of related figurations sustained by interrelations of meaning (rather than by the structural linkages that make up, for example, figurations of figurations). Such discourses are not just rational constructions.¹³ First and foremost, they establish ‘affective bonds’ () or, as we would prefer to put it, relations of meaningful interdependency, across different figurations, and in a variety of ways. Let us think about the web of figurations in which the figuration of a contemporary family is involved: that family has the responsibility of organizing the leisure and upbringing of its children; it interacts over many years with organizations (schools, adult education centres, universities) that are regarded as having certain responsibilities for education. But over time, those children develop independent relations with external institutions (cultural institutions for leisure purposes; companies and organizations that are responsible for providing employment), and so the family over time develops meaningful relationships with an ever-expanding, indeed changing, set of other external figurations (and figurations of figurations), right up until the parents become old and in need of care. Those external figurations are themselves further understood to stand in relation to various levels of government in a public domain, which media institutions themselves help to define and shape.

Our argument is not that such webs of figurations connect up in such a way as to create an overall functional whole of societies. Our approach is precisely not functionalist, since relations of meaning within such webs are always potentially contestable and interruptible. We want instead to insist on the importance of such connecting discourses whereby the ‘macro’ becomes embedded in the specificities of ‘micro’ actions through relationships of meaning. Put another way, individual actors come to learn to act in a social world characterized by discourses about what ‘society is’ and how ‘society works’ ().

Such relations of meaning can emerge in another way too, not directly through discourses of meaning but through deep practical relations of interdependency that ‘hang together’ as what seems to be a ‘way of life’ that, as such, gets taken to be meaningful ‘as a whole’. So for example actors in an economic market depend on the workings of the transport system, and actors in the financial markets depend on the workings of communications systems. If one breaks down, the other cannot continue. Such relations of interdependency continue up to higher dimensions without limit. Such is the complexity of the practical interdependencies on which contemporary life depends that actors struggle to make sense of all of it as a meaningful whole: this is where new types of discourse or myth emerge to make sense of it all, as a whole, as a reality ().

What we call ‘society’ is much more than a container for different figurations (); rather it is the overall ‘hanging together’ of figurations (and figurations of figurations) across the many domains of action associated with a large spatial territory. Insofar as these forms of hanging together are based on meaning, they have a mythical rather than material character. Those myths do not describe an ‘objective reality’ of the social world, but are particular constructions of it, which help hold in place assumptions about how its domains of action ‘fit together’ in a wider ‘order’ (see Chapter 10 for more details). There is a long history of such myths, which have been associated with media institutions and their ongoing claims to social legitimacy, and this is not the place to unpack this history in detail. But it is enough to mention that today’s myths of this sort go far beyond the legitimacy claims of old media institutions (for example, major public media), and encompass the mythical claims about collectivities brought together on social media platforms and the access to the ‘social’ supposedly achieved through ‘big data’.¹⁴

Figurations and Social Transformation

In the process perspective we are taking here, no social phenomenon is just given. Even figurations that remain over a long time ‘the same’ – figurations of religious organizations, for example – have to be constructed as ‘the same’ through action and interpretation. Alongside change, there is also ‘inertia’, yet attempts to preserve figurations and to avoid change can have the unintended side-effects of ‘strengthen[ing]’ a figuration’s ‘tendency to change’ (). Good examples are religious organizations or political parties in which the efforts of the organizational elite to keep them stable stimulate tendencies of critique, long-term instability, and so change. And equally it is possible that practices which are intended to change a figuration might strengthen its tendency to remain unchanged. For example, some actors’ ideas of change might be oriented in different directions, and by this make specific changes difficult. Any description of figurational transformation on a large scale is necessarily complex.

What weight finally should we give to media technologies of communication in such transformations? It is a question of the ‘moulding’ potential that derives from a figuration’s media ensemble. A changing media ensemble in a family, in a peer group or in a shared office community does not necessarily transform the figuration itself: the family, peer group or office community might stay the same. A changing media ensemble only results in transformation if a figuration’s practices of communication are also transformed and with them the ways in which meaning is produced within that figuration.

Nor is internal change of the media ensemble simply driven by external changes in the media environment. In many cases where figurations transform – maybe in most of them – it rather will be as a result of an interaction between internal and external forces. Think of a company: very often the reason to change its media ensemble does not take place accidentally but by management decisions. New internal data systems or new social media platforms are introduced to fulfil certain aims better: to process information more efficiently, or to reach customers more effectively. The media ensemble is made to change because at least some of the actors hope to better achieve their aims. But the side-consequences may be very different from how these actors intended it. Employees might, for example, appropriate data systems in subversive ways to build up their work procedures in a more convenient way. And whether changes in a media ensemble change the dynamics of a figuration remains a complex matter for local investigation. So, as already noted, some figurations (religious organizations) may remain quite stable in spite of their increased uses of digital media: the fact that the office of the pope is active on digital platforms does not change its power-balance as an institution. In each case, it is a matter of understanding the dynamics of particular interrelations. There is no ‘logic’ of media that drives changes in figurations: there is at most media-related change of various sorts in the field where the relations that make up a figuration, or figurations of figurations, unfold in a non-linear way.

Both internal and external aspects of a figuration can transform fundamentally in a changing media environment. This has first to do with new possibilities of how figurations can become related to each other. As we have argued in Chapter 2, the fundamental characteristic of each medium is that it offers possibilities to act beyond the here and now, and by this to extend the reach of human agency. It is, as Elias already noted, especially through media that ‘chains of interdependence become more differentiated and grow longer’, and so ‘become more opaque and, for any single group or individual, more uncontrollable’ (). This means that figurations can more easily spread across space and time (something we will discuss in the next section of this book). A company, for example, can much better integrate dispersed parts of its organization; administrations can reach different places; a family can hold together while being at the same time located transnationally. This does not mean that location ceases to matter in an age of deep mediatization. On the contrary, our relations with the media manifold privilege locations of high media connectivity (): across such locations figurations can spread much more easily, and come into contact with each other, for example the transcultural environments of cities that offer opportunities for linking actor-constellations in multiple ways (). It is a key characteristic of huge cities and metropolises, that these are places where very different figurations cross and intertwine.

Life with media in the age of the media manifold is inseparable, we have argued, from involvement in a variety of different figurations that exist in complex and sometimes contradictory constellations. Such figurations exist in arrangements of varying levels of complexity, involving, on the one hand, links between actors or the incorporation of pre-existing figurations within ‘figurations of figurations’, or, on the other hand, relationships of meaning that in practice bind particular figurations (and figurations of figurations) into large-scale interdependent relationships. So far, we have just described the ‘how’ of figurations, and we have not yet reached the question of ‘so what?’, that is, the consequences of figurational patterns for wider social order, or what we might call the ‘figurational order’ of contemporary societies. We reach this finally in Chapter 10, but before that, in the preceding chapters of Parts II and III, we need to pass through various intermediate stages of analysis: in Part II, space, time and data; and in Part III, the implications of deep mediatization for the self, for collectivities, and for our possibilities of governing social space.

The Distinctiveness of Simultaneity in an Age of Deep Mediatization

We noted, in the last chapter, media institutions’ historic role in sustaining ever larger collectivities’ sense of togetherness and practical coordination across space. Because of the interlocking of space and time, we could just as readily insist that media institutions have for a long time played a key role in sustaining spatially bounded collectivities in time (). There are some distinctive features of how today, with deep mediatization, media and information infrastructures sustain relations of co-consciousness (as Schutz would put it) across space and time. First, they do so between any one point and any other, since the building of internet access into many mobile phones and devices means that simultaneity need not be orchestrated only from an institutional centre of production: I can draw you and others into continuously tracking me simply by uploading to a digital platform a picture of what I have just done in a different continent, although in doing so I am of course relying on an infrastructure that depends on many networked concentrations of resource. Second, transmission speeds are close to instantaneous, whatever the size of content transmitted. This does not lead to the erasure of time as experienced, but, quite the contrary, to a growing sensitivity to the ‘little temporal differences’ (): we notice (and often judge negatively) when an SMS or chat message goes unanswered for some hours, or when an email goes unanswered for a day. Third, there seems in principle no limit to how much the size of transmitted content can go on growing, even if, at a certain point, speed of transmission becomes for most purposes instantaneous, which implies that the burden of interpreting and processing content received from co-conscious interactors can also grow without limit.

Two decades ago Paul Virilio prophetically expressed the resulting superfluity of information as ‘the generalized arrival of data’ (). The relative acceleration of digital information transmission means that, rather than actors relying on moving themselves or their objects across space, they may stay in the same place and access information about most things where they are: the importance of the space in-between seems erased (). But the outcome is certainly not a universal sense of acceleration. Not only do people react against acceleration (as they perceive it), but intensified flows of communication lead to breakdowns of figurations and new forms of inertia or slowed-down reaction (). Meanwhile, our access to the past is changed by the expansion of information available in and across ‘the present’. Does this undermine the uni-linear nature of temporal development, as Rosa claims ()? We would prefer to say more modestly that we inhabit a social world characterized by the pluralization of temporalities, on the one hand, and the complexification of technological systems for the coordination of temporality, on the other.

Certainly, the range of interlocking forces pushing the social world towards ever greater temporal coordination – often locally felt as acceleration – is formidable: competitive economic pressures, cultural pressures, and socio-structural pressures, each of them intensifying one or more moments in the underlying ‘circle’ of seeming social acceleration. That ‘circle’ itself, according to Rosa, can be broken down into three elements: technical acceleration, acceleration of the pace of life, and acceleration of social change and the perception of social change. The key point is that the circle of influences that reinforce the experience of social acceleration is ‘largely immune to individual attempts to interrupt it’ (). Put another way, time is involved not just in the relations through which figurations, and figurations of figurations, are built, but also, more widely, in the construction and sustaining of a wider figurational order (see Chapter 4) which is not directly accessible to individual adjustment. Figurational order (to recall Chapter 1) involves large-scale ways of organizing things in time (meta-processes), which no actor can challenge without undermining the practicality of cooperation itself (something no one wants to undermine). Media and information infrastructures therefore achieve more than ‘technical acceleration’, the accelerated transmission of information: they actually shape the figurations through which these intensified relations of interdependence are enacted, and so the possibilities of social order through figurations.

The outcome can seem to be a dramatic temporal dislocation: a world in which the demands of work and system, reliant on mediated systems of communication, spill out far beyond the normal boundaries on which everyday habits seem to be built. There are serious issues for work-related inequalities here, and more generally the result, some writers argue, is to damage the fabric of life itself. Media devices are one of the most vivid embodiments of the push towards constraint connectivity and 24/7 living. In Japan, rules of instant response around the friend-focused platform Mixi have taken the ‘culture of instantaneity’ () and the principle of constant connectivity (found in many cultures today)⁹ to a high pitch of intensity: as a result, many young people sleep with their phones on or under their pillow (). Jonathan Crary writing from New York generalizes this phenomenon into a paradoxical ‘24/7’ way of living which, because it can never be realized (human beings have to sleep in the end, or die), imposes an impossible injunction whose effectiveness ‘lies in the discrepancy between [any actual] human life-world and the evocation of a switch-on universe for which no off-switch exists’ (). The crucial point however is that these processes are not simply processes of acceleration; they are instead intensifying figurations of meaningful interdependence whose medium is technologically based platforms of communication. That is why our sense of being ‘harried’ (under continuous pressure of time) is associated not so much with having less time, but with the problems of coordinated performance linked to the ‘density of social practices’ ().

Yet it is only apparently a paradox that these developments are described, and not just by marketers, as a form of freedom. For, if we understand freedom as a complex social achievement of collaborative interdependence, then as Claus Offe once noted, ‘the more options we open up for ourselves, the less available as an option is the institutional framework itself with the help of which we disclose them’ (). The injunction towards constant connectivity is part of that ‘institutional framework’. Because in democratic societies, at least, we start by resisting the idea that we are unfree, it requires considerable work to expose this authoritarianism to view, as something we must confront and change (). Elias’ concept of figurations however helps us see this infrastructure as a process of emerging interdependence that actively shapes our very possibilities for action and imagination. At the core of such processes of interdependence are changing forms of media and communications.

Media and the Changing ‘Speed’ of Social Development

Before we elaborate that broader point, let’s make a more basic point about how we understand the mediated construction of time. Time-relations are socially constructed. The point is not just that what counts as time is socially constructed (), but that what is measured together as continuous time is also constructed. According to Roberto Cipriani, ‘[t]he question of time is centred on a series of relations’ in terms of how two or more events are conceived as constituting some sort of series (). Cipriani quotes Elias: ‘clocks certainly help us to measure something: nevertheless, this something is not exactly time, which is invisible, but something which is very tangible like the length of a working day or the eclipse of the moon, or the speed of an athlete who runs the 100 meters.’ The spatial extensions of our social world through social media platforms also change the durations and sequences that are considered measurable (the ‘time’ of a Facebook newsfeed is one very recent, but now pervasive, construction of sequence).

In addition, the individual experience of time is constructed through the figurations in which we are involved. With the change of media and information infrastructures these figurations have also changed, transforming the temporal dimension of the social world and the positioning of the individual in it. Media, through their role in enabling new, ever more stretched-out, figurations of figurations, have changed the ways in which (the gearing through which) particular forms of accelerated process are interlinked with other forms. This affects the overall speed at which our social sense of time seems to be transformed.¹⁰

In Elias’ analysis, a new tempo of the social world developed historically over several generations, emerging gradually through felt contradictions between prior habitus and emerging social pressures: such a contradiction might occur over time, over the life course of each individual, but the social response only takes shape in new cultural guidelines that require generations to develop (). But with the embedding of norms from social media platforms into daily life we may be witnessing a faster transformation, over a single decade, with implications for intragenerational as well as intergenerational relations ().

Not that those changes are felt evenly across space; everything depends on the organization of practice. As some sociologists note, ‘temporal scales are institutionalized through the production and reproduction of [. . .] practices’, for example in organizations and professional environments, generating distinctive ‘temporal orientations’ (). In the next section on media’s detailed consequences for changing forms of temporal order, we need to bear in mind both universalizing pressures and how people are differentiated precisely in terms of their time practices, and their relative resources for controlling those practices.

Time Deficits

In some media cultures there is a growing sense in everyday experience of a deficit of time: not ever having enough time to do what one has to do. The problem is much more than technological acceleration: it is a matter of the changing interrelations between economic, cultural and social practices, affecting actors of all sizes and on every level, and media and information infrastructures’ changing involvement in all those interrelations. Indeed, because the internet is a connective space – its potential to connect is effectively unlimited – it amplifies this problem, without limit.

The individual actor may sense a contraction of the present. What we call ‘the present’ is not an objective measure of time, but our delineation of the everyday as a sphere of action and planning during which we are entitled to assume ‘no further change’. Accelerated communications change the ‘present’ because they create pressures to bring forward the moment after which further change must be accommodated. This is the consequence not just of the acceleration of communications, but also an ‘accelerated transformation’ in ‘the context of communication and action’ (). Only within ‘the present’ can we draw conclusions based on experience ‘to date’ and so securely orient our actions; when the volume of signals incorporated into ‘the present’ increases, perhaps to an arbitrarily high level, social actors may have a problem, and may lose the capacity to react to communications.

An important way in which this happens is through ‘multi-tasking’. The changing distribution of work across space increases possibilities for ‘still doing’ multiple tasks even if one has moved away from the location originally associated with that task. But if multi-tasking is facilitated by intensified communication-at-a-distance, it has a profound impact on our sense of ‘the present’. It imports the time-signals and time-related obligations from multiple activities into a single time-flow. Little wonder then that one of Southerton and Tomlinson’s UK respondents speaks wistfully of ‘the most relaxing part of the day [when] I only have to do one thing’ ().

Absolute comparisons of speed of transmission and interaction are therefore only a small part of the story we need to consider. After all, speed only affects us if it requires an adjustment in our practice, and such a requirement is only registered at all if it appears within the matrix of practices in which we are engaged. Many increases of absolute speed are of no phenomenological concern to us, because they are black-boxed within our experience, or occur in domains of activity where we are not direct actors (for example, the increased speed of electricity distribution or of weapon launching). Where acceleration does concern us, our feeling for it depends on at least two things: the intensity of activity required from us and – depending on the first – the intensity of others’ interactions with us that result. Since however our activities are generally meshed together within confined domains of, ‘time–space packing’ () imposes limits upon the resulting acceleration of interactions, and may act as a break on how much faster accelerated possibilities for interaction actually feel to social actors.

While there are only so many bodies of a certain size that can fit into a finite space – there are certain natural limits to spatial packing, beyond which the attempt to pack just has to stop (otherwise, bodies get crushed) – the same is not true in time: there is literally no limit to how many messages, each sent in a non-synchronous mode, can ‘be there together’ in one’s inbox, each requiring response ‘now’ across a range of communicative platforms. The situation is very different with white noise, where countless signals cancel each other out so that nothing distinct can be heard. The challenge of communication overload is that each message can be heard – as the carrier of a distinct meaning – yet it cannot be attended to, since the time required for doing so is lacking. In this way, contemporary arrangements for communication tend to generate time-packing demands on individuals, from moment-to-moment, which along with the related of communicative obligations they can never, in principle, fulfil.

Such multiple and impossible demands might not be problematic in ‘thin time’ where there is no wider normative framework for ordering action-sequences relative to each other. But they are deeply problematic in ‘thick’ time, or what Robert Hassan () calls ‘network time’, that is, ‘digitally compressed clock-time’ in which the temporal calibration of obligations within particular figurations is intensified. The contemporary workplace and the social relations of those periods of intense change in one’s social networks (such as adolescence or early adulthood) are likely to be periods of ‘thick time’ when the burden of communicative obligations left unfulfilled due to time-deficits is felt more strongly (). Problems of coordination in periods of ‘thick time’ become potential problems for any wider figurational order.

Practices in Response

So far our analysis has been on a general level, but we need to approach these possibilities also from the perspective of particular practices. A number of practices (simple and complex) offer entry-points into such transformations.

One practice in everyday life, which seems to point in the opposite direction from time-deficits, is ‘time-deepening’. By referring to time as ‘deep’, we mean here not to assert an actual dimension of time – time is not literally deep – but rather the experience, generated by the progressive acceleration of our obligations to interact, of having more to do, more ways in which we must be ‘adequate’, and more conflict between those increased obligations and our ability to meet them in the available time.

One digital media practice relevant here is archiving. We are, through digital infrastructures, archiving all sorts of information, images and other traces of life-processes more easily. As a result, photography becomes embedded as a social practice in new ways (). The wider implications are complex: greater institutional capacities of memory require improved means for interpreting and sorting the now vast piles of information that accumulate. Meanwhile, the difficulty of ruling out the possibility that some past incident will have been stored in some more or less embarrassing form – so that it can be released by someone at some future point – increases the risks for individuals and institutions of managing that uncertainty.

We may also feel time-related obligations to each other of a more complex sort. Think of today’s frequent expectation of individuals that they will keep all channels open (), already hinted at in the discussion above of the problems of constant connectivity. We can now, if we wish, be permanently open (and potentially responsive) to content from all directions. Many writers see the practice (or even compulsion) of continuous connectivity as characteristic of the media generation that grew up with digital media (). Enabling us to be open on all channels in this way is part of the marketing promise of new portable interfaces such as the smartphone. While it is impossible to be open to everything, the demand to ‘be available’ shapes an emerging practice that is different from earlier modes of media consumption based on intermittent communication and a clear distinction between mass media and interpersonal media. Keeping all channels open means permanently orienting oneself to the world beyond one’s private space, and the media circulated within it. It is against the background of this (previously impossible) standard that some people seek to limit their openness to communication, at least for certain purposes

In response to the new intensity of time-challenges received through media, we are developing practices of selection: processes by which we peremptorily stop doing certain things we always used to do; processes by which we drastically select from the environment with which we must interact in order to make it more manageable. Sherry Turkle’s striking account () of some of the drastic ways in which young people select out (de-select) the activity of just talking with their friends, because of the acute time–space-packing problems they face, is just one example of how the texture of everyday life may change quite drastically through indirect pressures of selection. In response to such pressures, we seek ways of ‘selecting out’ from our communication environment while maintaining the illusion that we are still fully connected. Selecting out is increasingly delegated to technological interfaces such as the smartphone, which offer gateways to media that are the result of intense prior selection. By choosing from a vast range of ‘apps’, people screen out much of the infinite media environment and create a ‘chosen’ interface that is both manageable and seemingly personal: this is the double level of the ‘media manifold’ in action. There is the potential here for our experiences to become fragmented,¹¹ but new forms of linking across disparate sites of experience are also developing. We are now sharing aspects of experience – images of special meals we have just eaten, selfies that record our presence at a location or with a person as we travel – in ways we did not do before.

One important selection practice is ‘hiding out’. ‘Hiding out’ (being online while trying to disguise this from others, or avoiding using the phone for its primary use, speaking) is increasingly common in the USA. Sherry Turkle quotes a 21-year-old college student: ‘I don’t use my phone for calls any more. I don’t have the time to just go on and on’ (). This, as Turkle sees, creates a paradox: infrastructures for enhancing interpersonal communication, through their built-in tendency to accelerate interaction, create time-deficits so severe that people have to stop communicating, at least directly (face to face), deferring ‘full’ communication so as to manage their time-deficit better. Hiding out then is part of a wider set of ‘practices of demarcation’, where people mark off space–times when they will not be connected to certain individuals, collectivities or organizations (): on holidays of course, but also other ‘slots’ in the daily flow of patterned time (). Such demarcations impose further adjustments on others: the costs of those adjustments may be absorbed within a wider system of managing mutual availability, or their disruption may fall unequally on particular individuals or particular classes.

Time and Social Order

There is also a broader point about coordination. The problem is not just one of lacking time for reaction to communications, but lacking time for interpretation, that is, for making narrative sense of what one is supposed to be up-to-date with. The problem may not be amenable to direct adjustment. We reach here a wider problem of figurational order: of ‘configuration’, as the phenomenological philosopher Paul Ricoeur put it. In Ricoeur’s work, the relationship between time and narrative is explicit: ‘narrative’ is only ‘meaningful to the extent that it portrays the features of temporal experience’, while ‘time’ only ‘becomes human time to the extent that it is organized after the manner of a narrative’ (). The possibility of narrative for Ricoeur always requires acts of ‘configuring’, a temporal synthesis which grasps together a variety of heterogeneous elements. Narrative offers a different way of being in time, since the plot ‘extracts a configuration from a [mere] succession’ (), the mere succession from moment to moment. There is always, for Ricoeur, a paradoxical relation in human life between the possibility of configuration and the reality of mere succession.

Ricoeur’s reflections on time and narrative were developed principally for a philosophical and literary context: social ordering was not Ricoeur’s priority. But they remain a useful reference-point for thinking sociologically about our lived experiences of media and information infrastructures, and in particular for the problems of ‘configuring’ the much intensified form of these infrastructures that we now experience: extracting a configuration (of possible narratives) from an endless stream of mere succession. Ricoeur himself was well aware of the historicity of the narrative structures within which we operate:

perhaps, in spite of everything, it is necessary to have confidence in the call for concordance that today still structures the expectations of readers and to believe that new narrative forms, which we do not yet know how to name, are already being born [. . .] For we have no idea of what a culture would be where no one any longer knew what it meant to narrative things. ()

Ricoeur intuits a future vulnerability of culture with respect to time, which we are only now, and in particular cultures, beginning to register in practice: the possibility of a new ‘culture’ which resists narrative and creates conditions where the configuring of individual experience is only partly possible.

Problems of figurational order – or of ‘configuration’ in Ricoeur’s term – through which the interrelations in which we are enmeshed make sense to us as an order are an expression of the growth of new figurations, and figurations of figurations, that constitute those interrelations. And figurational problems felt at the level of individual actors are a manifestation of the emergence of new forms of generalized order by which the world can be governed in new ways. We saw in the last chapter how a growing complexity of spatial reach, made possible by new technologies of transmission, enables new worlds of interaction, such as the trading floors of global stock markets. In this chapter’s final section, we explore other ways in which order might be emerging in time across today’s increasingly densely woven mesh of social figurations.

Time and Work

Time is important for work since ‘work is done in time: it is a temporal act, done by actors’ (). The management of time within the bounded spaces of organizations is one of the key dimensions and problems of organizational life (). But work environments may differ markedly in how they are organized in time. Lee and Sawyer draw a key distinction between monochronic and polychronic work environments:¹²

Individuals working polychronically place less value on temporal order, accept events as they arise and are likely to engage in multiple activities simultaneously. In contrast, people working monochronically seek to structure activities and plan for events by allocating specific slots of time to each event’s occurrence. ()

Most organizations, because they focus on common systems and goals, assume a monochronic way of working, even if individuals operating within them also operate polychronically: that intensifies the normalizing pressures of time structures. Maintaining a common time-environment for working becomes more difficult in distributed work that, in turn, is facilitated at a general level by technologies of communication-at-a-distance. Sarker and Sahay () discovered this when they studied the work of virtual teams developing information infrastructure projects in the USA and Norway. Their research was focused on the role of time and space in shaping the practices of individual team members, and what they found was interesting:

Key problems related to time appear to be arising from mismatches in psychological and social clocks of team-members, complexities in accounting for time zones, negative interpretations of time lapses, and difficulty in comprehending temporally disordered sequences of chat and threaded messages. ()

Such work involves developing effective forms of coordination and collaboration between people who may not have worked together before, or even seen each other before. A particular issue is ‘suspicion’, that is, the ‘difficulty in physically verifying the actions of remote members’ (). In response, people may overcompensate, as evidenced by a study of Australian home workers:

I feel if I don’t answer an email someone thinks I’m purposely ignoring them instead of I haven’t read it yet. It’s a concern and it’s also just how I see myself as a professional. I want people to know I am looking after things. ()

A further problem is communicative ‘silence’: a time-period without communication in what was supposed to be continuous communication. People at a distance tend to interpret communicative silence negatively, as caused by ‘incompetence or a lack of commitment’, sometimes leading to ‘a breakdown of even functional relationships’ (). Ways of dealing with this include minimizing the dependences between locations – scaling back the degree of collaboration over distance – or developing new norms of communication (for example, an assumed 24-hour response period). To sum up, ‘while ICTs act as key enablers of distributed work’ within a technologically mediated space–time, ‘they by themselves do not guarantee “location transparency”’ (). Translocal communication creates complexities of managing the temporal sequence of communications that may undermine the mutual substitutability of perspectives that Schutz regarded as necessary for effective social interaction. The pressures towards reliance on translocal cooperation meanwhile accumulate, meaning that the costs of such imperfect working conditions, at the individual level, can be high.

Consider also the organizations that make the information packages, which form much of the background of our shared time: the ‘news’. Recent research with major global news producers shows that a change in the inputs to news is affecting those producers’ relation to time. Schlesinger and Doyle () argue that, whereas the temporality of news-making was once clear (focused on ‘breaking news’ and its onward, outward transmission), a changing economic model has meant that news producers must increasingly adapt to the inward flow of audience response and commentary (for example, what is trending on Twitter and Facebook). This leads, they argue, to real-time adjustments in production routines which potentially introduce time-conflicts into the practice of news-making (): do you spend time checking or identifying a source for a new story, or do you spend time checking social media reactions to your last story?

Such paradoxes of time-relations through media affect informal types of work too. Think of those outside conventional organizations who are trying to change the economic system: protesters for social change. As anthropologist Veronica Barassi () points out, protesters, like others, have to deal with ‘the temporality of immediacy’: that means face-to-face interactions and activities must regularly be interrupted by incoming electronic messages, and the need to send out new messages. Barassi’s fundamental point, similar to ours, is that it is ‘through the organization of our everyday human practices that we construct specific temporalities’ (). It is not just a question of the personal cost of being continuously available on one’s mobile, but also of the trumping, for everyone, of certain time-consuming types of activity by others which seem more ‘immediate’. The quality of political practice, especially practice aimed at producing social change through extensive deliberation, may thereby be damaged, as this activist interviewed by Barassi notes: ‘I feel that you cannot create a real discussion [on social media]. The communication is too fast, there is no depth. It is also difficult to establish a history of events and thoughts.’ Or as another activist puts it, ‘we need also to propose our alternatives. The problem is that these complex analyses need to be developed properly, we need time and space to do that’ (). Other researchers of social movements note the cost for activists of being locked into ‘an [accelerating] event-oriented dynamic’ that focuses on the following up, and responding to, social media trends (). The apparent obligation to ‘stay connected’ creates hidden time-deficits that, in turn, generate costs for the wider practices in which actors are engaged, without denying however that digital media also facilitate practices of memory and archiving in political movements ().

Broader Social Orderings through Time

What general principles can we draw from these disparate cases? First, there is a tendency for the temporal dynamics of communication systems to override the temporal dynamics of other processes in which the actors receiving the outputs of those systems may be involved. Social media platforms seem to carry their own sense of time and time-related obligation, as a number of commentators have noted (). Political activists may feel a pressure to follow only the spikes in social media (what is ‘trending’, for example, according to Twitter’s algorithms), rather than register a steady growth in interest in their activities, even though it is the latter that may be more useful and sustainable (). The fact that this message is ‘right now’ on one’s phone – and therefore immediately available to be responded to – seems to trump other possible uses of the time required to respond. There are parallels in the world of general social interaction: the fact that a message from an assumed close consociate on the Mixi platform generates a requirement for a young person in Japan to respond immediately (), rather than perform some other action. Because responding immediately, while doing many other things, can be difficult, an overarching requirement may emerge to sustain a state of constant readiness to respond whenever a message comes in, even while one is asleep. There is something paradoxical when technological system imperatives, in the temporal mode, trump other major needs; for example, for continuous periods of non-responsiveness (usually called ‘sleep’). How system imperatives are integrated, well or badly, into everyday reality becomes crucial for the quality of that life.

The problem here for analysis is arguably more difficult than that of grasping the order of complex social domains (such as global trading rooms) built on many interlocking figurations oriented to shared communication and data flows. For those domains carry their own narratives of order and, although its reality must be lived by those workers, we can make sense of the idea that such domains operate within boundaries that sustain their effectiveness. More difficult to analyse are cases where system imperatives linked to generalized communications infrastructures bleed out into daily life for individuals, and where narratives to make sense of the resulting disruptions are unavailable. This is the largely uncharted area that Elias captured through his notion of ‘tempo’.

In Elias’ terms, a dramatic recalibration of the social world’s tempo – considered from the perspective of some actors relative to others – is under way, and from the perspective of deep mediatization social media platforms are driving it. Elias says that ‘tempo’ works as ‘a function situated at a junction of so many chains of action [which] demands an exact allocation of time; it makes people accustomed to subordinating momentary inclinations to the overriding necessities of interdependence; it trains them to eliminate all irregularities from behaviour and to achieve permanent self-control’ (). The normative force of tempo doesn’t derive from anyone’s intentions: it accumulates through interlocking mutual relations of many figurations, and figurations of figurations, which over time result in what seem to actors to be ‘overriding necessities of interdependence’. The assumed reason for this ‘overriding’ is that cooperation will break down without them. When one type of call on our time is trumped by another, this has consequences: if this trumping becomes routine, this may cause us permanently to direct scarce time away from the uses that have been trumped, opening the gates to an exponential growth in the time-use that generates the trumping. This is one important engine for the progressive reshaping of everyday life in time through media. To the extent that we lack ways of making sense of this change (of configuring it with our other ways of making sense of the social world), a problem of figurational order arises.

Here we see at work one aspect of deep mediatization that is driven not by any ‘logic’ inherent to media contents or forms, but by the dynamics of the intensified interdependences of meaning and sociality that media make possible. Yet this is only one dimension of how social media platforms are working with deep mediatization. Also important are new modes of evaluating others, commenting on what they have just said or will shortly do, imitating others, all of which have temporal aspects, and so contribute to increased temporal interrelatedness. The possibility that our relations with time are changing through what we do on digital platforms has been noticed by various writers (). We are in the middle, potentially, of a major transformation in social ordering whose outcome will depend on more than temporal calibration. As Elias understood so well, questions of tempo cannot be separated from questions of value.

The Transforming ‘Here and Now’

One hundred and fifty years ago, a speeding up of the movement of bodies and things (railways) disrupted society’s ‘traditional space–time continuum’ (), leading to the loss of the ‘spatial distance created between localities [which] was the very essence of their “here and now”, their self-assured and complacent individuality’ (). It is too early to tell whether the ‘consciousness’ (as Schivelbusch put it) associated with the here and now of everyday locations, as it was understood before the advent of continuous and media-based connection, will be lost completely, but we should not underestimate the convergent forces pointing in that direction.

A banal, indeed parochial, example can speak to the sort of problems emerging in many places. In the UK in September 2015 it was reported (Guardian, 4 September) that an unnamed 14-year-old boy who flirted with a girl by sending her a naked picture of himself is now recorded in a police crime database, and will remain so for at least a decade. He sent the image, presumably not thinking of its archived after-life (he might not have expected that the recipient would be quick enough to circulate the image on to others). Nor did he expect the image to come to the notice of a police officer based at the school who recorded the ‘crime’ in the database, although without making any charges. The result of this ‘negligence’ (and the original no doubt foolish and offensive act) is however an ongoing penalty that seems wholly disproportionate. Something in the figurational order seems out of joint when communicative practices in space (circulation) and in time (archiving) generate consequences whose scale is radically at odds with those anticipated by actors themselves.¹³

An additional force at work is the huge investments involved in converting the interconnected space–time of everyday communications into a domain for profit: however uneven access to communications remains, the push to ‘connect’ a very large proportion of the world’s population, for example, by cheap ‘smart’ phones or ‘free’ internet provision such as Facebook’s Free Basics platform, is real.¹⁴ As Jose van Dijck () explains in her masterly survey of the growth of so-called ‘social media’ platforms, it is impossible to separate this trajectory and its purported transformation of ‘social life’ from the development of software which organizes the ‘data’ of our interactions into a space of appearances for us as social users. There would be no ‘social media’ platforms without that process of organizing data, which is not to deny that data processes in themselves already have social consequences. One consequence of the embedding of data-based processes in everyday life may be to shape the reference-points by which we organize action. Those who design data-processes are particularly interested in prediction. When the predictive results of such data-gathering are fed back into our own streams of experience (‘personalized’ adverts, differentiated prices when we buy something, newsfeeds on our social media page, prompts to take action, commendations on our Twitter performance), social actors’ own sense of time – the various linked time-horizons towards which their actions are oriented – may get changed in the process.

We will return to these questions of order, and their implications for questions of value, in Chapters 10 and 11. We need however in the next chapter to consider more closely a theme that has emerged across Chapters 5 and 6: that is, the implications of ‘data’ for phenomenology’s account of how the social world is constructed.

Databases and Social Classifications

Crucial to this transformation is the database; ‘the ability to order information about entities into lists using classifications [is] a contemporary key to both state and scientific power’ (). The database has a distinctive type of power which Bowker defines as ‘jussive’, an ordering power based on an ‘exclusionary principle’ that determines what can and cannot be stored in a particular form (). The consequences of database operations are in this sense final: ‘what is not classified gets rendered invisible’ (). The point of database operations is to fix the starting-point (the base) from which data operations – counting, aggregating, sorting, evaluation – begin. In that sense, by being placed in a database, ‘data’ becomes ‘unmoored’ from the underlying materials from which it was gathered (). It makes no difference to this process that some of the processes developed on the base of the underlying data architecture may themselves go on adapting in response to emerging patterns in data, provided the structure of the underlying database remains the same.⁸ Such adaptations are in any case driven not by some independent ‘will’ present in the processes from which data are gathered but by the emergence of features that, from the perspective of the data process, are defined as significant enough to trigger such adaptations. In all such cases, the ‘knowledge’ that is produced cannot be separated from the purposive selections out of which the database is formed, or subsequently adapted. Insofar as the outcomes are treated as direct knowledge about the processes re-presented by the data, they are misleading. As Bowker put it, ‘our memory practices [are] the site where ideology and knowledge fuse’ (). We have already in Chapter 3 suggested that these developments can be seen as a potential new fourth wave of mediatization, through which our interdependencies progressively deepen through infrastructures for the continuous production and exchange of data: the emerging wave of datafication, within the wider wave of digitalization.

Most data involved in the organization of social life are ‘made from the raw material of human experience’ () and used for social classification. The purpose of data-gathering is not neutral, but precisely discriminatory, that is, ‘to coordinate and control [people’s] access to the goods and services that define life in the modern capitalist economy’ (). This discrimination requires a massive pooling of computational resources: a single database is not sufficient. It is essential to aggregate separate databases into massively larger ones, enabling the matching of patterns across countless sites of data collection, from which predictions can be made (). From the development of ‘distributed relational databases’ in the 1990s and 2000s to the work of today’s growing data sector, including global businesses on the scale of Google, Expedia, Acxiom (), the aggregation-for-value of data originally collected from specific locations and in specific contexts now comprises a basic fact of social life.

It is naive to ignore the consequences that the principles of selection underlying such data-gathering may have for particular distributions of power, and for the long-term organization of the social world. As Theodore Porter put it in a broader context, ‘quantification is a technology of distance’, motivated not by a ‘truth to nature’ but by ‘the exclusion of judgement’ () that makes particular types of judgement possible and efficient through a reshaping of the social world itself: ‘the quantitative technologies used to investigate social and economic life work best if the world they aim to describe can be remade in their image’ (). Indeed, because large-scale institutions necessarily act at a great distance from the realities they seek to influence, the operation of data processes plays a crucial role in making a distant, intractable ‘world’ into an ordered, calibrated reality that can be interpreted and governed: as anthropologist James Scott put it, ‘legibility’ [. . .]. is a central problem in statecraft’, requiring a ‘politics of measurement’ (). But, as some legal theorists have noted, we must extend this analysis to the whole surveillance apparatus of the contemporary commercial and information technology environment ().

We miss the social form of this complex process, unless we connect up the abstractions inherent in data functions to the experiential processes in which those functions have become embedded (), and so register the potential violence of data-based ‘processes of [. . .] representation and classification’ (). For this, phenomenological reference-points are essential.⁹ But a core feature of data infrastructures is deeply at odds with the assumptions of classic phenomenology. That is its opacity:¹⁰ ‘the configuration of networked space [. . .] is increasingly opaque to its users’ (). Data’s continuous processes of selection and comparison have a generalizing force across an infinite domain that guarantees a larger asymmetry in social knowledge. As Christine Alaimo and Jannis Kallinikos () put it, ‘it is exactly because of the abstract nature of data and the simplicity of the logic of encoding that the social can be represented in all of its (now compatible) highly pliable forms [. . .] Once the social gets engraved into data, it ceases to be related to established categories and habits. It is transposed onto and thus enacted according to the very same logic’ by which it was generated. Or, as Jose van Dijck put it more succinctly, ‘it is easier to encode sociality into algorithms than to decode algorithms back into social action’ (). The translation of social life into ‘data’ therefore casts a large shadow: the domain of descriptors that fail to get captured by data processes ().

The result is a change – a potentially profound change – in our relations to infrastructure. Infrastructures are, at root, tools for human action, operating at the highest level of complexity, a black-boxed substrate of ‘ordinary’ human action. In the digital world, our infrastructural tools (for example, our pages on a social media platform) are increasingly entangled in powerful and distant processes, which we cannot unpick or challenge. All tools involve mechanisms whose details we forget when we use them, or perhaps never knew: we may guess how a hammer has been put together, even if most know little about how the modern car is constructed, yet both hammer and car are tools of everyday living and our ignorance of their workings (their black-boxedness) is not crucial to the quality of our use. But many of today’s ‘digital tools’, as we use them, are black-boxes of a different sort, black-boxes that are also in the act of using us.¹¹ They track our actions algorithmically, not to enable the tool to work better for us, but to generate data for the toolmaker’s use: that is, to enable us to be better targeted by advertisers and marketers (). This is indeed the very rationale of the much-hyped ‘Internet of Things’, yet its transformation of our usual relations to infrastructure seems not to have been noticed.

The result is a social relationship to abstraction very different from that envisaged by Schutz. Schutz saw artefacts as comprising the extreme end of the spectrum of ways in which humans typify (abstract from) their world (). But today’s data-based artefacts now themselves operate to typify humans mostly for commercial ends and surveillance, to construct a seamless world for commerce and control. We might call this tool reversibility. Tool reversibility is not immediately apparent when we use data-based tools, but it becomes apparent through our practices of use and the obstructions those practices encounter: whevever we use a data-based tool, it is already using us. This is one of the deeper cultural and social implications of the embedding of algorithms in everyday life ().

Categorization

Data processes rely, in turn, on categorization. Categories have been important in social theory for more than a century. For Durkheim and Mauss (), categories (as outputs of a system of classification in so-called ‘primitive societies’) were derivatives of the actual divisions of society itself, and of the very idea of society itself. In most subsequent accounts, the order of causality is reversed with categories contributing to ‘the built information environment of a society’ (): ‘typification’ playing a similar role in classic phenomenology. But to grasp the degree of abstraction involved in data-based categories, we need to look more specifically at their features.

As David Berry explains, no process of computer-based categorization (and so no sorting, combination or evaluation based on it) can operate unless an ‘object’ has been created: ‘in cutting up the world in this pattern, information about the world necessarily has to be discarded in order to store a representation within the computer [. . .] those subtractive methods of understanding reality [. . .] produce new knowledges and methods for the control of reality’ (). So, in order to compose the objects in a database, such abstraction, first, needs to have occurred. As already noted, there is no raw data, but only ‘data [. . .] produced through measuring, abstracting and generalizing techniques that have been conceived to perform a task’ (). Second, where large numbers of objects are to be processed by automated functions or algorithms, processing requires prior organization, the design of a database structure ‘to extract the data located in them as rapidly and as effectively as possible’; in this sense, ‘a data structure forms a sort of intermediate level, an abstraction mechanism, in the process of addressing machine memory’ (). Third, the more complex the operations to be completed, the greater the need to combine data levels and so enable more complex processing, what Fuller and Goffey call ‘abstraction layers’: ‘the more that different functions of a process [. . .] or software can be integrated by one layer of implementation, the wider it circulates and coalesces. The more generality an abstraction is capable of, the greater its degree of usefulness; and the greater its tenacity in self-stabilization, the more activity is arrayed around it’ (). In this way, more and more of what was once heterogeneous information can be processed together, but at the price of ever-greater levels of abstraction. Fourth, the processes of calculation performed on data must be automated through the use of algorithms. Algorithms are often identified, casually, with the whole process of data-based transformation of everyday experience, but as ‘encoded processes for transforming input data into a desired output, based on specified calculations’ (), algorithms are merely one of many elements in a sequence of progressive abstraction, if an essential one.

Alongside the relations between data processes (involving categories that do social work), we also need to consider what happens when the outputs of those data processes are played back to social actors. That they are played back () is another factor, which a sociology of knowledge that takes data seriously must grasp. Categorization¹² is fundamental to all forms of organization, including social organization. Without it, effective (or at least non-random) interaction with the world would be impossible. Yet in the social realm, categorization has a distinctive feature neglected in phenomenology’s account of typification and not even conceivable in Durkheim and Mauss’s society-driven model. As Ian Hacking pointed out, classifications of human objects are ‘interactive’ in a way that, arguably, classifications of non-human objects are not:¹³

Ways of classifying human beings interact with the human beings who are classified. [. . .] classifications do not exist only in the empty space of language but in institutions, practices, material interactions with things and other people [. . .]. people are aware of what is said about them, thought about them, done to them. They think about and conceptualize themselves. Inanimate things are, by definition, not aware of themselves in the same way. ()

Hacking’s insight has particular importance for the digital age when actors and actions are relentlessly categorized in countless ways for various purposes.

Our ways of interacting with categories are not easy to disentangle. They occur not randomly but in a highly structured context linked to the purpose for which data is being gathered in the first place. The simplest example is the social media platform. As Daniel Neyland notes, algorithms do not have a simple or automatic recursive effect on the social world, but involve a ‘configuration through which users and/or clients are modelled and then encouraged to take up various positions in relation to the algorithm at work’ (). What is striking about the configurations of social media platforms is that we act on and through them, largely as if there were no such configuration: indeed the very idea of ‘platform’ is a constructed space where the interface between everyday interaction and commercial transaction appears natural, a seamless dataflow (). Through our underlying desire to maintain our social commitments across the newly configured spaces where they seem to migrate, there emerges a ‘growing social commitment to functionality’ (Plantin, Sandvig et al., forthcoming). This has major implications for how the social world is constructed.

Translating Data into Practice

There are at least five fundamental ways in which data abstracted from social experience can translate into frameworks for social practice. These connect with the dimensions in terms of which the other chapters of Parts II and III are organized.

The first relates to the organization of space.¹⁴ As Kitchin and Dodge () analyse extensively, many spaces (physical, organizational, informational) are now ‘coded’. Their operations are structured through the software that processes data inputs of various sorts: the highly controlled space of the airport security queue is one clear example (), entry into which is impossible without having met various data-related conditions in a prescribed sequence. This is an aspect of the rise, more generally, of the automated management of social processes. Unlike traditional surveillance, this form of control allows no gaps, since it operates through ‘a grammar of action’, that is, ‘a systematic means of representing aspects of the world [. . .] and an organized language for processing those representations’ (). Under these conditions, the spaces Kitchin and Dodge call ‘code/spaces’ are figurations (in our term) of a particular, highly organized sort, driven by today’s complex forms of interdependency.¹⁵ Social media platforms feel like ‘spaces’ where, quite simply, we encounter others, but their existence is shaped by the underlying operation of platform software and its calculative infrastructure. Insofar as they create publics, these are ‘calculative publics’ (), not that calculability itself is new: Weber () already saw this a century ago as ‘the peculiarity of modern culture’, but it has a constitutive role today that is unprecedented.

A related point can be made about the time of online media. Online media encourage us to operate in a distinctive time of required reactions related to the ‘expected’ rhythms of platform interaction: the Facebook timeline, the Twitter hashtag stream (). This time is not natural but the result of configuring time-sequenced data in a particular array designed to stimulate ever more interaction. We relate to this array as if it were a natural production by the human parties involved in the exchange, yet without the data-based presentation of the platform, there would be no mutual orientation in space–time, and so no ‘interaction’. Many platform devices (such as email reminders) are designed to train people to rejoin the flow of what has been called ‘social media time’ (), should they slip out of it. This helps stabilize new data-based figurations that can function as ‘social metronomes’ ().

The third translation operates at the level of the self. Each of us is familiar with the need to operate as a self under various descriptions: difficulties arise when contradictory descriptions of ourselves converge on a single interaction, and this much is familiar from social phenomenology. Unfamiliar from that classic work is the idea that each of us has not just a self-based identity (vis-à-vis the state or corporation with which we must deal), but a constantly updated ‘data double’ that is the resultant of the vast data stream that each of us generates continuously across various sites of data-tracking (). The ‘data double’, with its built-in relation to multiple interdependent systems of data capture, depends entirely on the ‘standardizing [of] classification systems so that they are comparable and databases can be joined up’ (). It poses sometimes difficult challenges for individuals whose overall data stream may generate undesired or conflicting data, and, as already noted, the ‘data double’ is cut off from the shadow body of data that does not – perhaps cannot – feed into the relevant processes of calculation ().¹⁶

The fourth translation operates at the level of collectivities. Data classifies, and so data processes not only work to specify individuals uniquely, but also generate countless groupings to which individuals are treated as belonging (). Whether these groupings correspond to anything that might be recognized by social actors as collectivities outside the process of data generation is an open question, but we are familiar with cases where the insistent use of data labels generates a type of action. Think of Facebook ‘friends’: some of these will have been friends before but many others are likely to be those one has acquired through the practice of receiving and making friend requests on the platform. As Taina Bucher puts it, ‘friends have become a primary means through which the production and occlusion of information can be programmed’ (). In everyday actions and adjustments, actors become attuned to maximize such data-based groups (another example would be numbers of Twitter followers). Collectivities are sites where, through data processes, new norms of action and reaction emerge. We already know however that data processes are creating new entities for governments and civil society actors to deal with. So when, during Brazil’s Junior Masterchef competition in October 2015, offensive and abusive sexual comments emerged online about a 12-yearold girl contestant, an NGO worker created a hashtag #primeiroassédio, which quickly generated more than 80,000 similar stories across multiple platforms (). The importance of open hashtags as attractors to form political action has been noted for a number of major global protests, for example the rise of the 15-M and indignados movements in Spain in 2011 ().

The final translation operates at the level of organizations and order and flows directly from the third. A problem for governments and corporations from the vast proliferation of continuous data streams is monitoring what counts as ‘risk’. According to geographer Louise Amoore, governments are increasingly relying not on judgement or deliberation (no longer possible perhaps in the face of such a vast mountain of ‘information’) but on an ‘ontology of association’ that ‘draw[s] into association an amalgamation of disaggregated data, inferring across the gaps [. . .] to derive a new form of data derivative’ (). According to Amoore, this new ‘politics of possibility’ involves a fundamental abstraction operating on the flow of time itself: it is oriented to the predicted future, to ‘a population yet to come’. ‘Data’ must be rendered ‘actionable’ (), which means selecting and excluding, ‘rul[ing] out, render[ing] invisible, other potential futures’ (). These practices of exclusion become the basis for governing and ordering whole territories, a point to which we return in Chapter 10.

In all these ways, the domains of the social world, their practices and knowledge are reconfigured, in part, through processes of categorization based on data. The distributed complexity of these processes is a key contemporary example of the figurations of figurations and other large-scale relations between figurations that we argued a materialist phenomenology must understand. The emerging figurational order that results from this transforms the basis on which, in specific social domains such as the family and school, we are bound into figurations of interaction. It also operates, potentially, to change the very stuff of the social domain that powerful actors such as governments see themselves as acting upon. In the next section, we consider some implications of this for how individuals and collectivities act in the social world.

Monitoring

Although online platforms themselves direct surveillance, the new monitoring affordances of digital communication technologies involve us all: from the basic question (‘why hasn’t s/he texted yet?’), to Googling others before we meet them, to more persistent forms of mutual monitoring, via multiple forms of ‘social media’ which include not only social networking sites, but also sites for posting ‘user-generated content’, ‘trading and marketing sites’, and ‘play and game sites’.¹⁷ For that reason, legal theorist Helen Nissenbaum avoids the term ‘surveillance’ because of its strong pejorative associations with the state, and proposes the more open term ‘monitoring’ (). Indeed the ‘digital trail’ is a major factor in the life of children, who in a country such as the USA are subject to continuous monitoring by their parents (). Self-monitoring and self-tracking () is another important part of the picture, sometimes with specific goals (a sick person willingly accepting a measuring device that can warn a local hospital of symptoms of an impending heart attack), but often for more diffuse purposes. Emerging here in everyday practice is what Jose van Dijck calls ‘the ideology of dataism [. . .] a widespread belief in the objective quantification and potential tracking of human behaviour and sociality through online media technologies’ (). In some institutionalized fields such as health, practices of continuous monitoring are a new and urgent trend; in others, like education, they build on decades of increased measurement and surveillance in schools (). Indeed a consequence of the increasing capacity to combine vast database networks with huge calculative power is that ‘the more data there is, the less any of it can be said to be private, since the richness of that data makes pinpointing people “‘algo-rithmically possible”’ (). The result over time may be a certain fatalism. We may come to accept a social world characterized by continuous and enhanced mutual monitoring as our starting-point for thinking about the social. If so, this is a new and clear example of what we have called deep mediatization.

Berger and Luckmann’s principle that ‘human “knowledge” is developed, transmitted and maintained in social situations’ () – that is, situations where human beings, by virtue of their mutual dependence on shared resources, must come together to act and think – now carries a very different implication. The wave of digitalization (Chapter 3) has created a continuous plane of interaction based in technologies of mediated communication where, in principle, any actor, wherever located, can reach, and be reached by, the communications of any other. The temporality of social situations (Chapter 6) has also been transformed in more subtle ways, giving access to aspects of the flow of daily life that were previously lost, once experienced. The resulting enrichment of experience is inseparable from a new degree of institutionalization of social form. As Jose van Dijck () puts it: ‘through social media, these casual speech acts [of previous everyday life] have turned into formalized inscriptions which, once embedded in the larger economy of wider publics, take on a different value’. Let us think a little more about the implications of this.

Social situations, through their increasing involvement in ecologies of measurement and counting, are deeply implicated in data’s status as a source of economic value. While many aspects of the metricization of social space are hidden to social actors, this cannot stop data-processing becoming entangled in the emotions of everyday life: ‘more than mere tools, algorithms are also stabilizers of trust, practical and symbolic assurances that their evaluations are fair and accurate, and free from subjectivity,error, or attempted influence’ (). Translated into the language of classic phenomenology, algorithms and other aspects of the data infrastructure become a form of ‘objectivation’, part of ‘the process by which the externalized products of human activity attain the character of objectivity’ (). That is why an exposé of the searcher-sensitive features of the search engine results in which we trust is shocking ().

For Berger and Luckmann, institutionalization depended on ‘the generality of the relevance structures’ achieved in the production of knowledge (). The interoperable metricized space of social media platforms and online interaction generally are coming to comprise, with little resistance, a new structure for generating social knowledge. According to a recent survey, of the US parents who use social media, 75 per cent go there for advice about how to solve parenting problems (). Put another way, in the language of US pragmatism, the ‘generalized other’ () that regulates social action is now increasingly sustained by a commercially encouraged flow of online exchange. Gillespie notes that ‘algorithms impinge on how people seek information,how they perceive and think about the contours of knowledge, and how they understand themselves in and through public discourse’ (). If the algorithms associated with particular platforms, sites and practices have acquired legitimacy, anyone and any organization that depends for its power on legitimacy must deal with the consequences of ‘what appears’ somewhere in the unbounded, linked space of the internet. The management of this ‘new visibility’ () becomes an all-consuming challenge, creating new challenges for the self: ‘being spotted “by accident” against one’s will is not an option; missing out on purposeful display becomes a predicament’ ().

An interesting example in the private corporate sector is the transformation of the hospitality industry, and its relations with customers and employees, through data-based systems for collecting customer reactions. As Orlikowski and Scott () note, the force of platforms such as TripAdvisor (popular at the time of writing) is striking. In an industry whose principal asset is customer anticipation of a good service, a new mechanism of legitimacy in the ‘space of appearances’ is a profound shift, especially when the results of such recommendations are now distributed ever more efficiently as data within the corporate sector (). For political organizations that must sustain legitimacy through a narrative of control of their past but also their ability to manage a whole society’s future, the dynamics of legitimacy and information are even more complex (). New forms of interdependency are emerging here, based not on digitalization but on datafication, and the link of datafication to categorization. ‘People put things into categories and learn from those categories how to behave’ (). Whether or not social actors are aware of the many levels of data processing that work to shape their contexts of action, they interact with data-based contexts such as social media platforms as if they were sites for social categorization and normalization.

It might seem, therefore, as if a hurricane is blowing through the domains of the social world and our knowledge of them, threatening to overturn every reference-point and previously bounded context of knowledge production. Two important factors constrain this chaos, although they do not in the process protect the social from datafication, and from the potential normative problems that it generates. First, there is the stabilizing force that, as Berger and Luckmann put it, ‘institutions do tend to “hang together”’ (): meanings or (as they put it) ‘relevances’ overlap between contexts and institutional settings, constraining divergent interpretation, establishing norms of comparison, and (we would add) generating criteria for ensuring things do not appear. Arguably, this will work to entrench processes of datafication into social reality through a contemporary version of the circular feedback between structuring forces and structured outcomes that Bourdieu called ‘habitus’.¹⁸ Second, there is the ongoing and irreducible tension between the grain of social experience and the forms in which it appears online. Social actors are likely to devote increasing efforts to contesting this tension with varying degrees of success.¹⁹ We return to the implications of these struggles for social order in Chapter 10.

If we take seriously the possibility that the automated digital tools that measure behaviour and activity online are now a key part of everyday life’s background, then phenomenology has been complicated irreversibly. A materialist phenomenology must register how everyday actors are involved in bringing the workings of those tools into their everyday awareness. All are processes of categorization. As Bowker and Star note, categories create a ‘social and moral order’ (), but in the case of data-related categories, it is unlikely to go uncontested.

The figurations of figurations that make up the distributed data industries and the domains that rely upon them are transforming the space of social action. In this chapter, we have given key emphasis to the role of data in social media platforms. There is a reason for this: this is precisely where the process of constructing social reality is remoulded in detailed forms. As Kallinikos and Constantiou () say, ‘social media platforms elaborate architectural arrangements through which communal interaction and daily living are transformed into data ready to enter the circuits of calculation and so-called personalization’. But social media platforms are only one area where data processes are becoming deeply embedded in the building-blocks of social action. Others are the growth of data-generating ‘wearable’ devices in the health sector, and this is just part of the wider ‘Internet of Things’, whose consequences for the texture of the social world are, at this point, uncertain.

We can put all this in a broader philosophical perspective. The philosopher John McDowell () considers how our ‘mindedness’ (our unfolding conscious relation to the world as human beings) becomes embodied in forms of interaction and resource, as part of what he calls our ‘second nature’ (the evolving set of social institutions which humans are disposed to develop alongside their first, biological ‘nature’). For McDowell, the forms that embody the ‘possibility of an orientation to the world’ have a history (),²⁰ and this history is constantly open to revision. This formulation enables us to frame with particular clarity the problem that data pose for the social world as conceived by classic phenomenology.

Berger and Luckmann assume that the forms of our ‘mindedness’ evolve only from the accumulation of sense-making by human social actors, but what if today there is an alternative ‘embodiment of mindedness’ ()? What if ‘data’, in all their direct and indirect forms, are being installed as an alternative and exterior cognitive infrastructure through which not only do we become minded, but the world becomes mindful of us, and everything we do? Since the data processes discussed in this chapter are part of an informational infrastructure that is being spread globally at huge speed, this amounts to a further stage in the deepening of mediatization. The very scale and scope depends upon the delegation of knowledge generation and knowledge application to automated processes. Once delegated, those processes become exterior to the process of social knowledge as classical phenomenology conceived it: they become what we might call, building on McDowell, a ‘third nature’, driven by the economic imperatives of the data industries and all the wider goals of capitalist expansion that, in turn, drive those industries. This ‘third nature’, if it is to order social life, requires social actors to adapt to it in a process that, following Agamben (), we can call ‘subjectification’: the production of entities that can function as subjects within this new type of social order. One achievement of materialist phenomenology is to remind us that we are those entities – unless, that is, we refuse to be.

In sum, in the wave of datafication, new means for producing social knowledge have emerged with two key features. First, they produce ostensibly social knowledge through automation that is necessarily exterior to everyday processes of human sense-making. Second, they are oriented to goals, driven by wider economic forces, that are different in type from the goals that embodied actors are able to have, unless, that is, they give up on their autonomy entirely. The result is the emergence, unevenly at this stage, of a new kind of sociality – call it ‘computed’ or ‘platformed’ () – that changes the starting-points for everyday reflexivity and sociological reflection. Social order (which classic social phenomenology set out to explain) is now, through its very conditions of formation, inhabited by a form of already ‘rationalized reason’ (),²¹ which cannot be comfortably integrated into the reflection of individual social actors. Positive readings of this world () are possible, but they cannot cover over the fissure developing within the production of social knowledge itself. Not surprisingly, in response, some call for the ‘right to disconnect’.²² The only repairs to this fissure must lie in the agency of social life on various levels, to which we turn in Part III.

Putting Digital Traces in Context

But how can we understand digital traces in detail? Digital traces are more than just (big) data: they are a form of digital data which becomes meaningful only when a sequence of ‘digital footprints’ is related to a certain actor or action, typically (of ) a person but in principle also a collectivity or organization. It is the tie of data to the unique individual that underlies why marketers and other institutions of data processing are highly interested in collecting and aggregating data. Such data is not just any information, but always information linked to processes of counting, as reflected in the French expression for digital traces, traces numériques. Digital traces are numerically produced correlations of disparate kinds of data that are generated by our practices in a digitalized media environment. Because of the tie that such correlations always have to a certain entity or process in the social world, digital traces are a major, but not the only, reference-point for how a social entity acquires its ‘digital identity’.

This deep mediatization of the self has, unsurprisingly, generated much debate about how sociology can, any more, stay in touch with what selves are and do.¹⁴ Some academics go one step further and argue that the self’s digital traces amount to more than traces of a self that still (in principle) can be reached through other routes (observation, listening). They argue that such traces offer for the first time a direct access to ongoing processes of social construction. Maybe the most prominent example is Bruno Latour’s integration of the investigation of digital traces into his overall approach to social analysis (). A ‘digital traceability’ () then becomes a possibility for analysing processes of social construction in situ: ‘being interested in the construction of social phenomena implies tracking each of the actors involved and each of the interactions between them’ (). This approach attempts to move beyond the old micro/macro divide (a classic topic in sociology)¹⁵ by arguing that statistical methods allow us to get at ‘macro’ phenomena directly by an analysis of the individuals’ online activities. With digital traces, they claim, we have a direct access that allows us to witness processes of assembling in the moment they take place ().

This argument is, in our view, fundamentally mistaken: it misunderstands the nature of digital traces in relation to the self. The error begins with misinterpreting the social world as ‘flat’ and so accessible to analysis simply through an aggregation of trace-patterns registered in the various data domains that are underpinned by media infrastructure. Such patterns may well have some value, but this move reduces the complexity of the contemporary social world to a flat plane without differentiated levels, so replicating on a large scale the problem with the concept of assemblages that we have critiqued already in Chapter 4. Second, and even more fundamentally, such an approach misunderstands digital traces as something ‘neutral’, offering us a ‘direct access’ to the social world. However, they are not ‘neutral phenomena’ but rely on the technical procedures of governing institutions that produce this kind of information. The construction of the self’s traces therefore already, through its very process, inscribes certain interests as well as visions of society: we find here a strange echo of Jacques Derrida’s () insistence on how subjectivity (‘the self of the living present’) is always ‘a trace’ that links to a ‘temporalization’ and ‘spacing’ that carries beyond itself. Digital traces do not offer access to the social world ‘as it is’ but an access to the procedures whereby powerful organizations attempt to construct a world on which they can act.

A materialist phenomenology must therefore take a rather different approach to the self’s digital traces. Far from treating them as direct ‘traces’ of what is, it insists on approaching them from two directions: first, in terms of their consequences for the everyday world of the individual and, second, in terms of their origins in the world-making strategies of governing institutions. Because all social classifications are ‘interactive’ (), these two aspects overlap in the flow of practice, but at no point do they converge into the possibility that the self’s data traces provide direct access to the social. This complicates our evaluation of contemporary movements that attempt to transform reflexivity through processes based precisely on datafication, to which we turn next.

Self-Quantification

Quite how big an adjustment data practices demand of sociology’s traditional approach to the self emerges when we return to the original thoughts of Berger and Luckmann on self-consciousness. As they put it:

The world of everyday life is not only taken for granted as reality by the ordinary members of society in the subjectively meaningful conduct of their lives. It is a world that originates in their thoughts and actions, and is maintained as real by these. [. . .] we must attempt to clarify the foundations of knowledge in everyday life [. . .] [in] the objectivation of subjective processes (and meanings) by which the intersubjective common-sense world is constructed. ()

To some extent at least, this starting-point is uncontestable, but in certain key respects it is now in tension with a new notion of self-consciousness and self-knowledge that is emerging through the automated collection of data. Gary Wolf is a key proselytizer of the ‘quantified self ’ movement: while aware of the tensions and potential strangeness of the idea that ‘self-knowledge [comes] through numbers’ (), he also provides some of the key arguments in apparent support of this proposition. For example, the idea that ‘our ordinary behaviour contains obscure quantitative signals that can be used to inform our behaviour, once we learn to read them’ (), which of course builds on the practical starting-point discussed throughout this chapter that ‘social media made it seem normal to share everything’ (ibid., added emphasis). The process is self-fuelling: ‘the more [people] want to share, the more they want to have something to share’ (). This is an explicitly collaborative form of mediated construction that, as it grows in regularity and intensity, produces alongside each concrete individual a ‘data double’ () which – from certain perspectives like those of the quantified self movement, but also perhaps some contemporary governments () – contains more ‘truth’ than an individual’s own self-reflections. We are witnessing here innovations in relation to the fundamental languages for describing and measuring the self: we cannot therefore imagine them to be innocent of power. On the contrary, what is under way is a transformation of social and political power: as Julie Cohen () puts it, ‘we are witnessing the emergence of a distinctly Western, democratic type of surveillance society, in which surveillance is conceptualized first and foremost as a matter of efficiency and convenience’.

The case of self-quantification in the health domain is particularly illuminating. More and more people are using tracking devices to generate continuous data about themselves (for example, heart-rate, metabolic rate and so on). As Deborah Lupton points out, this is more than just another ‘technology of the self’ (): it is a way of embedding the self, and its ‘data practices’, in a much wider infrastructure of data generation, aggregation and analysis, which potentially might transform the distribution of resources in the health industries away from cure and towards continuous activities of prevention. Aside from the installation of self-measurement into the basic processes of everyday life (a ‘personal Taylorism’ (), this changes what counts as self-awareness, with so-called self-quantifiers using ‘data to construct the stories that they tell themselves about themselves’ (), and the norms that orient reflexive self-awareness.¹⁶

So far these practices are the preserve of a small movement of enthusiastic first-adopters, but they have behind them a considerable momentum and leverage.¹⁷ Some in the health industries see patients’ continuous sharing of their health data as part of an (increasingly obligatory) practice of ‘self-management’ (). Though sometimes disguised as a game, as a sort of play, self-monitoring ‘enrolls people into self-governance by using their highest aspirations and capacities, that of selfcase and self-development’ (): once we acknowledge the gaps in our normal memories, why not supplement them with the ‘more objective’ materials generated automatically by continuous data collection ()? Yet, the price of playing along is acceptance of a data collection and data-sharing infrastructure whose rules are non-negotiable (). There are of course problems like tracking devices, which may generate too much information to be interpreted (). But those concerns may easily be overridden by more powerful claims: for example, that data collection tools enable subjects to become conscious of their otherwise unconscious behaviours and behavioural patterns (); that this is how individuals take responsibility for preventive medicine (); or more generally that this is how individuals optimize their ‘performance’ along various dimensions ().

Positive readings of these developments are readily available: as a protective ‘technology blanket’ wrapped around the self (); as an ‘improved “higher quality” self’ (); as a form of collaborative memory (Frith and Kalin, forthcoming) or ‘macroscope’ (); or even, according to one leading proselytizer, Kevin Kelly, as a form of an ‘exoself’ (). There is no doubt of the serious commercial intent to build such a comprehensive layer of data-tracking around the individual, at least in the context of the richest societies and their highly resourced health and personal development industries. To speak of such developments as a transformation of the ‘self’, however, precisely begs the key question: whether such quantification practices are compatible with other senses of the self and, even if so, whether the price they exact is too high. Once we leave aside the more crass rhetoric (for example of a ‘people-powered health’ at the Nesta web page),¹⁸ the notion of every person carrying an ‘algorithmic skin’ is closely tied to the futures of ‘commercial, governmental and medical research’ (), and mobilizes a discourse about self-awareness for distinctively institutional ends. As Williamson puts it:

Self-quantification produces a ‘calculable public’, a public that is presented back to itself through the data organized and coordinated by algorithmic approximations of its traceable health activities. ()

So strong is the drive towards greater self-awareness and more sustainable self-improvement that it is easy to lose track of the potential damage that is being done here to our everyday notions of self-consciousness which have never, until now, needed to resort to external data-gathering infrastructures to validate their claims about self, others and the social world. There is a risk already, even before we consider the biases that the process of metrification and datafication introduce into daily life.

Towards the Institutionalization of the Self

We have explored in this chapter a self that, through the operations of the media manifold, increasingly interacts in extended domains of the social world – which in turn act back on the self, increasingly through processes of datafication. This self is a site of tension between its own claims to awareness and new notions of ‘enhanced’ self-awareness derived from new automated data-gathering techniques. These potential new transformations carry a price. While apparently enhancing the freedom of the self, they build into the fabric of the self an infrastructural dependency: a process of institutionalization and materialization that, because of the asymmetrical power-relations at work in the media domain involved, introduces a dimension of unfreedom that today’s selves must confront ().

Deep mediatization – that is, the integration of media-based processes and relations into the very elements from which the self sustains its project as a self – introduces therefore a new friction into daily life. We will consider this problem more fully in Chapter 10. Undeniably however, if the site of the self is transformed by mediatization, then so too is the site of the ‘we’ that comes together through the grouping of selves. The risk that we mimic each other in certain ways and so converge artificially in our behaviours on online platforms, becoming ‘partial analogues of others’, has been noted for a while (). We need in the next chapter to explore more broadly how the construction of ‘collectivity’ operates in an age of deep mediatization.

Media-Based Collectivities

Media can constitute collectivities in two ways. First, they can offer by their content a frame of relevance for constructing such collectivities. Second, they can offer the space of communication in which these collectivities get constructed, regardless of the actual content that meet their specific frames of relevance. In the first case, media are constitutive in the sense of constructing the meaningful boundaries of these collectivities. In the second case, media are constitutive in the sense of supporting the communications practices through which these collectivities always get constructed. Each type has its specific dynamics that requires a more detailed analysis, but both need the label ‘media-based collectivity’.

The clearest examples of media-based collectivities are those that gather around particular media content (). An example for this, often discussed in media and communications research, is ‘audiences’, especially for exceptional media events (): people who follow television sports games, ceremonies, extraordinary popular shows or comparable ‘events’ that are communicated as a source of collective identification. Later we will discuss in more detail to what extent processes of constructing ‘imagined collectivities’ (of the nation or of other kinds) are at work here. The point we want to make is that, even if these people do not necessarily feel themselves to be part of a community, they may still build a more loosely connected collectivity as the spectators of a particular media spectacle ().

With reference to certain forms of media content, we can also witness the emergence of more stable collectivities for which ‘fan communities’ or ‘fan cultures’ are a prominent example (). Media are important here in a double sense: first, they define the relevance-frames for such figurations; second, they are important as means for keeping these collectivities together. With digital media the possible influence of these collectivities increased as new ‘politics of participation’ became possible ‘not simply through the production and circulation of new ideas (the critical reading of favourite texts) but also through access to new social structures (collective intelligence) and new models of cultural production (participatory culture)’ (). The digitalization of photography and the rise of platforms for the easy sharing of images (those created digitally, and digitized archival images) have enabled new collectivities to focus on sharing memories in new ways (): as one participant in Richard MacDonald’s study put it, ‘I’ve shown [my photos online] because it might jog someone’s memory’ (). While we have to be careful not to romanticize these collective cultures (), it is evident that digitization has expanded their scale, scope and regularity. Importantly, some media-based collectivities may now operate translocally, joining together Taiwanese fans of, for example, Japanese and other foreign television programmes, usually watched live through various unofficial online means (). But we have to be aware that using the word ‘community’ to describe them is not necessarily helpful. Quite early on there was a discussion about how far ‘interpretive communities’ () necessarily constitute groups of people who know each other and have a self-understanding as a group, or whether they might actually be much more loosely attached collectivities. Such a discussion regains relevance as the ‘new digital environment increased the speed of fan communication, resulting in [. . .] ‘just in time fandom’ (), partly experienced through digital platforms and ‘second screens’ in parallel to other forms of media use. The figurations of these fan collectivities become more diverse and ever more deeply related to media technologies. Therefore, instead of understanding each and every fan culture necessarily as a single community, we might do better to understand it as a complex figuration of figurations that links up different local groups in a range of interdependent activities.

Other media-based collectivities include various sorts of ‘online groups’, and again it is an open question how far they are communities. The owners of digital platforms especially have a tendency to call themselves ‘communities’ and understand by this rather a kind of forum function (). However, we should be careful not to mix such ‘technological definitions of “community”’ () with sociological ones. Basically, online groups are figurations that are built with reference to a certain platform and the topic of communication there. But it is an empirical question whether and how far these collectivities make progress towards becoming a community (Weber’s question about ‘communitization’). Contemporary digital platforms offer possibilities for creating a variety of different online groups on one platform, each of them based on various topics of interest, and various software add possibilities for online group-building; for example, multi-user online games where game-related collectivities like ‘guilds’ play against each other and whose construction is supported by the game software in various ways (). The ‘guilds’ may be built, for example, by text- or video-chat in parallel to the game played: such game-related collectivities may derive from or result in offline relationships, or remain solely online (). The degree of community involved depends on the individual case and its meaning.

Media-based collectivities can also be local and situated; for example, ‘flash’ or ‘smart mobs’. A ‘flash mob’ can be defined as a large group of people who gather by the support of digital media in some predetermined location, perform some brief action, and then quickly disperse (). The term ‘smart mob’ () originally having a more specific political focus, although the distinction has become blurred (). Whatever exact term we use, such mobs are forms of collectivity that have (digital) media as a pre-condition of their existence, and are a figuration tied to particular local gatherings or situations. In this they are similar to other new forms of situational collectivities, like ‘mobile clubbing’ (groups of people who go from bar to bar while connected by mobile media (), or ‘mobile gaming’ (), or ‘urban swarms’ of protesters (). Whatever their duration, these figurations, in their close relations to media, are typical of an era of deep mediatization.

Mediatized Collectivities

Even collectivities whose existence and formation are independent of media can form what we can call ‘mediatized collectivities’: families, peer groups, migrant groups or groups of excluded people nowadays are collectivities whose forms of meaningful belonging are, in part, constructed through the use of media. Here we find what Nancy Baym calls ‘networked collectivism’: ‘groups of people now network throughout the internet and related mobile media, and in-person communication, creating a shared but distributed group identity’ ().

When it comes to families, the appropriation of media – especially of television – was and still is important for maintaining them as a collectivity (). However, the crucial point here is that keeping up family life became a cross-media endeavour (). When family photos are shared on online platforms and through that a family memory is constituted (), or when family relationships are articulated by digital media use (), it is the whole media ensemble that is involved. Such a mediatization of the family enables new forms of the figuration of the family, especially families that are spread across long distances and at the same time keep up a close relation to their family members (): their ensemble of different media (mobile phones, internet-based visual telephony, email, texting, digital platforms) makes it possible to keep up family roles like that of ‘mothering’ across long distances. But media change the ‘feel’, ‘texture’ and ‘meaning’ of a family’s relationships: the relationships between parents and children, for example, if constructed by video conferences, telephone calls and mobile phone surveillance, remain more distanced than one constructed mainly in face-to-face interaction ().

Similarly with peer groups. While nothing new, peer groups, especially of young people, are nowadays moulded to a significant degree through their use of media, not least because mediated popular culture provides them with a relevant point of reference. Indeed, an increasing portion of young peer-groups’ communication as a group takes place via media: mobile phones, digital platforms, chat apps (). Members of peer groups feel under pressure to appropriate media and to fulfil the rules of communication specific to them (). Group membership becomes defined by access to particular media ensembles so that failure to use certain media may result in groupexclusion. Put differently, in the age of deep mediatization, membership of a peer group is enacted through appropriating its media ensemble.

Further evidence for the mediatization of collectivities comes from migrant groups. Nowadays the very act of migrating is already highly intertwined with media: the ‘image’ of the place to which one is migrating as well as the possible migration network is built up via the internet before the act of migration (). Migration itself is organized by digital platforms and smartphones, which together allow detailed navigation, ongoing information, as well as documentation of the migration process. The relevance these media have is closely related to the ‘information precarity’ () of refugees in large camps: without technological and social access to relevant information, with irrelevant, sometimes dangerous, information prevalent, unable to control the circulation of their own images, and under continuous risk of surveillance by state authorities. The ‘connected migrant’ () is involved in various mediatized collectivities during his or her journey, building mediatized groups of support along the way, and maintaining at the same time contact with family, friends and others at the place of origin.Media have always enabled migrants to maintain links to their wider migrant group through various, mainly ‘smaller media’ (). But with digitalization these possibilities significantly increase ().

Deep mediatization may also transform the experience of intensely marginalized groups. One striking example is homeless people. For a long time, media have been relevant as sources of entertainment and opportunities for communitization in shelter homes and elsewhere (). With the spread of digital media, homeless people in mediasaturated societies become regular users of digital technology, especially of smartphones (). Beside organizational matters, they use these technologies to maintain contact with friends and families and for collectivity building (). Their media use goes beyond self-representation as with homeless persons’ newspapers (). It is much more about keeping contact with and remaining part of ongoing collectivities while still living on the street.⁷

Some Emerging Principles

It is obvious that the specificities and possible transformations of the collectivities discussed so far cannot be related solely to media. Other processes are driving forces too: individualization (), globalization (), commercialization (). But having these further meta-processes of change in mind and comparing ‘media-based’ and ‘mediatized collectivities’, it becomes evident that collectivity-building does not dissolve into a single form of individualized network. Collectivities remain a meaningful unit of human life in times of deep mediatization, but through mediatization become transformed in a range of ways. Three points are striking.

1. Media contents become important resources for defining collectivities when media contents become the ‘topic’ around which those collectivities are constructed. This is especially evident in media-based collectivities such as fan cultures that are predominantly defined by a shared enthusiasm for a certain media content (a series, a genre, etc.). But it applies, too, to mediatized collectivities – families, peer groups – that appropriate various kinds of media in constructing their moralities, rules, boundaries and joint experiences. While such content a decade ago was typically communicated by mass media (print, film, radio, television), and access to it was more limited, today a huge variety of symbolic resources is accessible via online distribution. Therefore, the spectrum of possible collectivities has increased fundamentally.

2. Media are means for constructing collectivities, especially for online groups that constitutively rely on their online space of communication, but also for those peer groups and families that become related to the use of media like smartphones. Actor-constellations may be sustained across long distances and collectivity experienced synchronically at a distance, even under circumstances of intense mobility, whether of individuals or the whole collectivity. New textures of collectivity emerge through a variety of media ensembles as well as very different opportunities for constructing collectivities. Together a collectivity’s specific features and the communicative capacities of its media ensemble define its possibilities of transformation.

3. Media trigger dynamics in collectivities. It is less the single medium that matters here than the whole media ensemble,⁸ the dynamics of which can however vary hugely: having access to certain media may become fundamental for becoming a member of this collectivity, or media may affect the communication that takes place within collectivities (online groups, for example, are well known for their practices of ‘flaming’, rooted in the lack of co-presence between their members). Even in mediatized collectivities, the degree to which members are ‘always on’ has consequences for the quality of their communications.

Elaborating the original ideas of Hubert Knoblauch (), we can call a shift from ‘collectivities of pure co-presence’ to ‘collectivities of multi-modal communication’. By this we mean that, before the spread of today’s communications media, human collectivities involved co-presence, in which everyone knew each other, practices typically were shared, and core knowledges were distinctive of the whole collectivity. This is the conception of community found in classic writings about communities (). But, with mediatization’s successive waves of mechanization, electrification and digitalization, further kinds of collectivities gained relevance that we can call ‘collectivities of multi-modal communication’. Based on and shaped by a diverse media ensemble, less rooted in direct experience but in shared processes of mediated communication, these ‘collectivities of multi-modal communication’ become communities when they build up a ‘common we’ as well as long-term structures. However, an important characteristic of deep mediatization is the variable intensity of such collectivities, and the role that choices between media modalities (media options within the media manifold) play in the formation of distinctive collectivities (what makes them ‘multi-modal’). Far from a general switch-over to purely ‘personal’ networks, in an age of deep mediatization we see a more differentiated range of collectivities, in part because even older collectivities of co-presence have now become mediatized.

Imagined Political Communities

Originally, the nation as an ‘imagined community’ involved the idea of national public media as crucial to the construction of this imagined community. In his enlightening analysis, Benedict Anderson emphasized ‘the novel and the newspaper’ as ‘the technical means for “re-presenting” the kind of imagined community that is the nation’ (). In such a perspective the ‘development of print-as-commodity’ is the key to understanding the construction of a communicative space that offers the possibilities to imagine a ‘national consciousness’ (). Electronic media later supplemented this process – mainly radio and television – which gave ‘print allies unavailable a century ago’ (). In this way, processes of communication that allowed the construction of the nation were intensified.

However, it would be a mistake to understand this mediated representation of the nation as an explicit discourse about the nation as a political unit. Rather, it is a ‘banal nationalism’: a habitual representation of the nation as a point of identification in a ‘mundane way’ (). This process of construction works through how, in media, a ‘homeland’ is articulated as a ‘here’ and the group of the people living in this homeland as a ‘national we’. Contests and conflicts with other people become a competition of ‘nations’, and even the weather is something that is automatically related to a national territory. This ‘constant flagging’ of the nation ensures that, ‘whatever else is forgotten in a world of information overload, we do not forget our homelands’ (). And even today this process of constructing the world as a world of nations continues, for example in online platforms that are not necessarily bound to a national territory (). For various kinds of political actors – politicians, parties, governments, and journalists – the imagined community of the nation remains the point of reference for constructing social order. This keeps the imaginary of the nation as a ‘quasi-natural’ unit of living and identification: ‘It is a form of life in which “we” are constantly invited to relax, at home, within the homeland’s borders’ ().

With globalization in general and globalization of the media in particular, however, such social imaginaries became weaker (). Besides the ‘project’ of constructing the nation as a collectivity, other kinds of ‘projects’ of imagining collectivity became more widespread. One prominent example for this is the ‘community of Europeans’ that can be understood in parallel to the nation as a ‘community of communication’ (): it is imagined through collective processes of communication. Here, the underlying communicative space is a transnational and multi-lingual public sphere that emerges from the increasing discussion of European issues across borders as well as an increasing monitoring of European political affairs in Brussels (). While on the level of everyday experience this kind of imagined collectivity has not the ‘natural’ character of the nation, we can see under way an ongoing construction of a ‘banal’, however contested, Europeanness ().

Alternative Ways of Imagining Collectivity

But such alternative territorially related communities are only one way of imagining collectivity. With deep mediatization we have a variety of other publics and imagined collectivities that partly conflict with each other and partly connect to each other (). This starts with ‘personal publics’ () or ‘private spheres’ () grouped around certain individuals, and ends with the ‘networked publics’ () of digital platforms that are characterized by a particular communicative architecture which enables these spheres of communication (). Around some topics, situational ‘issue publics’ () emerge across various digital media, including the mobile phone itself (). What we can notice here is a massive differentiation and multiplication of the different spaces of political communication, shaped in particular by underlying inequalities of socio-economic resources. This makes essential a ‘context-centered model’ of media’s role in the formation of collectivities (). We do not even know therefore the full variety of imagined collectivities that such publics support, but we know they expand far beyond the confines or reference points of national states or confederations.

To explore this more closely it is helpful to consider the case of online blogging. What is called the ‘blogosphere’ () is an online space of bloggers who are more or less closely interrelated. Typically, these relations become visualized as networks of technical () or semantic links of (mutual) personal references (). The main question here is what kinds of collectivities are built by these bloggers. Partly, they are understood as a kind of ‘community of practice’ (), being preoccupied with a certain topic, referring more or less to each other, and so building up an arena of discourse.⁹ However, we need again to ask how far the term ‘community’ is helpful, or whether this collectivity is defined simply by the shared interest of the involved bloggers. The situation becomes even more complicated if we consider the bloggers’ readers. The ‘intense affective unification’ () of blogging on a certain topic can result in ‘online crowds’ with their own dynamics: members of this collectivity come together on certain online sites, imagining themselves as a kind of collectivity of political interest and expressing their political position in affective ways.¹⁰ With such ‘online crowding’ we are witnessing the unification and relative synchronization of publics in relation to certain political issues through shared affective practices () and through a structure of feeling ().

The multiplication of possible publics also multiplies the possibilities for constructing types of imagined collectivities. The most prominent examples of this are social movements. While social movements have a long-term history of imagining collectivity – the most prominent example for this is the international socialist movement – so-called ‘new’ social movements’ () like the environmental or alter-mondialization movement are characterized by their global imaginations of collectivity that move beyond any national or supra-national political units. With the support of media, these movements aim at transformations on a global scale (), offering new imaginations of collectivity based around shared ‘project identities’ () and offering ‘networks of hope’ (). However, there are good arguments to be cautious about such claims. Social movements certainly have better resources for collectivity building today than before digitalization: a prominent recent example for this was the ‘occupy movement’.¹¹ Yet, at the same time, the internet also offers political elites many opportunities to intensify and diversify the ways in which they sustain themselves in positions of power (). Therefore, the transformative potential of new political collectivities might be far more limited than their own imaginaries suggest. However, with digitalization the actual character of social movements changes. There emerges a tension between more loosely connected, individualized forms of political action on the one hand, and new ways of actually constructing political collectivity on the other. While the two seem contradictory at first glance, a second look shows that both are an expression of the changing figurations of social movements and their imaginations of collectivity.

In an important book W. Lance Bennett and Alexandra Segerberg describe this shift in social movements as being from a ‘logic of collective action’ to a ‘logic of connective action’ (). With digital media platforms, they argue, we can distinguish three kinds of social movement figurations: first, there is ‘collective action’ which takes place in figurations of ‘organizationally brokered networks’, characterized by a strong organizational coordination of action. Media technologies are used to manage participation and coordinate the organizational goals as well as the communication of other aims. Second, there is ‘connective action’ realized by organizationally enabled networks with a looser coordination of action: media technologies support communicative practices that enable more personalized forms of action. And third, there is ‘connective action’ that is supported by crowd-enabled networks with little or no formal organizational coordination. Here, we have a large-scale personal access to multi-layered media technologies and communication centred on emergent personal action. While this threefold distinction is certainly idealized, it addresses the diverse consequences of digital media for the structure of social movements today: digital platforms support both hierarchically organized social movements and a highly individualized political engagement that is more ‘me-centric’ ().

But we have to be very careful not to confuse this possible shift with a disappearance of imaginations of collectivity. Even much looser figurations remain dedicated to the construction of imagined (political) collectivities. In a careful analysis, Anastasia Kavada () demonstrated this for the Occupy movement, which involved protests without formal organization, supported by a very open figuration of activists linked by protest and media-based communications practices. In such an open figuration ‘social media followers formed an outer ring while the inner ring included activists who were participating regularly in the physical occupations’ (). Through the use of digital media platforms, it became possible to construct two kinds of collectivities: first, the collectivity of the group protesting in the streets and parks, and second, a collectivity that followed the events (a kind of imagined collectivity of supporters). Occupy became a movement with transnational impact in part because it offered symbolic resources for imagining oneself as part of such a wider collectivity, and in so doing supported its own spread beyond the figurations of local protests. Such more open structures of organizing protest do not result in a default to ‘me-centric protest’ but rather in a more varied imaginary of ‘protest collectivity’ (). All can find a place within this collectivity of those who define themselves as part of the ‘99 percent’ movement against global capitalism.

Collectivities for Media Change

Social movements are now aware how important media technologies are for social processes in general and for collectivity building in particular. As a result, they consider media and media infrastructures themselves increasingly as an object of political engagement. Roots of this can be found in the ‘alternative’ and ‘radical’ media movements of the 1970s () that aimed to achieve ‘alternative’ forms of public spheres (). With reference to network infrastructure and digital media, important examples are the ‘hacker movement’ and the ‘open source movement’. The hacker movement’s political aim was to make the implications of the increasing omnipresence of computers and datafication publicly known and thereby politically negotiable (). The focus of the ‘open source movement’ was first to foster a certain form of non-proprietary software development; later it became a political movement intertwined with a general political engagement in ‘open access’ to information: the ‘open data movement’ (). A remarkable hybrid example is the Chaos Computer Club, one of the world’s largest and Europe’s oldest hacker organizations (). More recent examples include the technologically driven ‘repair movement’, combining hackers’ competences with an engagement for sustainability and a zero growth economy (Kannengießer, forthcoming). Collectivities like these not only use media and their infrastructure to support their engagement for particular political aims; they consider media and their infrastructure as themselves an issue of political engagement.

This focus of certain movement collectivities on contesting media is a general characteristic of an era of deep mediatization: in such an era, media, and ways of reflecting on media, become part of the stuff out of which the social world is built and so, for example, larger collectivities come together as such. Prominent recent examples are the ‘media-related pioneer communities’ () like the ‘quantified self movement’ () or the ‘makers movement’ ().¹² While they call themselves ‘movements’, these collectivities are instead hybrids between social movements and think tanks. Like social movements, pioneer communities have informal networks, a collective identity, and a shared aim for action. More particularly, they come very close to ‘technology-oriented and product-oriented movements’ (), like the open-source movement (). However, pioneer communities are generally not involved in conflict-driven relations with identifiable opponents as social movements are: indeed, they are more open to forms of entrepreneurship and policymaking, lending them an affinity with think tanks (). Pioneer communities here share with think tanks an ability to produce ideas, and an effort to influence both public and policymakers.

The imaginaries that characterize all these collectivities are oriented to media technologies. The ‘quantified self movement’ imagines better forms of healthcare through collective self-measurement and the accumulation of personal and collective data. The ‘makers movement’ imagines that new technologies will allow decentralized forms of production and new collectivities of value creation that will supplant traditional forms of (industrial) production. Such pioneer communities spread their imaginaries transculturally.¹³

We could continue our analysis and discuss many further examples, whether religious collectivities¹⁴ or transnational political violence, which have become more and more based on building collectivity through media.¹⁵ But again the point we want to make is a more general one: that we are experiencing a shift in how collectivities are imagined for political ends. While the imagining of collectivity through media was for a long time predominantly related to imagining the nation, we are nowadays confronted with much more diverse and conflicting processes of imagination that can no longer easily be integrated into the container of well-integrated ‘national projects’. Looking back, the project of imagining the nation depended on a close fit with the waves of mechanization and electrification and with the nationally based media infrastructure on which they were based, whether we turn to France, the USA, or colonial Nigeria (). The high point of the communicative construction of the nation was in the second half of the twentieth century. With deep mediatization, and its more diversely configured media infrastructures, political projects for imagining collectivities become themselves more diverse, even contradictory.

Two points are striking. First, the communicative spaces of constructing these imagined collectivities are increasingly detached from territorial borders. This does not mean that the project of imagining national communities has come to an end. But such imaginations are increasingly confronted with mediatized ‘old’ and ‘new’ imaginations of collectivity: supranational imaginaries jostle for prominence with national imaginaries, while social movements offer new imaginations of transnational and transcultural political belonging, and new types of media-oriented collectivity emerge. A characteristic of deep mediatization is the existence in parallel of these conflicting imaginations of collectivity and the unresolvable diversity of political values and political projects that results.

Second, the close relation between our imaginaries of collective life and media becomes taken-for-granted. First steps in such a direction are the various social movements that focus on media and their infrastructures as a political issue. Maybe more characteristic for deep mediatization are the media-related pioneer communities which call themselves ‘movements’ but are more closely tied to existing power centres. Their imaginings of collectivity involve ideas of living better together, through and by media technologies: those imaginings are effected not just by circulating media, but by a much broader range of practices focused on media that constitute what Hilde Stephansen calls the ‘social foundations’ of publics (). In our terminology, media-related practices enable new types of publicly significant figurations to be formed. With deep mediatization these imaginings become concrete political projects, based in the material infrastructure of media and communciations.

‘Working’ for Brands and the Data Economy

We can trace this development back to the idea of ‘brand communities’. At first sight these are very close to the features of fan cultures, being collectivities built around certain (media) products or even (think of Apple) the producing companies themselves. However, it is worth looking more closely at this phenomenon. In their original sense brand communities were specialized, non-geographically bound communities, based on a structured set of social relationships among admirers of a brand (). They were ‘largely imagined communities’ () insofar as their horizon was not just the artefact of a certain kind of product (a computer or a car) but the brand that this artefact represents. The fundamental point about brand communities is that they cannot be ‘made’ by companies but are rooted in the everyday practices of those who use these brands (). Nowadays, such collectivities go beyond the companies’ own strategies of marketing and encompass the mediated communication of the members of the community. Such brand communities are not ‘sub’ or ‘fan cultures’ because their members see themselves not as something ‘special’ or ‘marginal’ but as general consumers with an interest for a certain kind of accepted brand (); even so, as with fan communities, the community-building itself takes place by interaction either at physical meetings or online ().

Building up brand communities became, paradoxically, a top-down strategy of marketing, which resulted in ‘brand collectivities’ with a far lower level of communitization. The possibilities of digital platforms stimulated companies to experiment with ‘online groups’ of consumers in order to foster relationships between and to them (). The parallel relationship that consumers have to a common brand creates already a figuration, albeit one characterized by an intense asymmetry. The hope of corporations is often to support the development of more reliable consumer relationships (), but, when they do so, something different takes place: the construction of a visible and therefore meaningful collectivity of ‘followers’ who do not necessarily perceive this figuration as involving ‘community’. There is a big difference between those who just ‘like’ the web pages of certain brands on digital platforms, and the brandrelated groups on these platforms founded by the users themselves (). The first group shares only a positive view of a brand, but is represented as a collectivity through data accumulation. In the second case there is an interaction from the bottom up between people who not only share an interest for this brand and its products, but also develop a joint discourse about it, and potentially a relation to each other (). Only the latter is close to the original idea of ‘brand communities’.

With deep mediatization and the expansion of mediated interdependence that it brings, this first kind of brand collectivity becomes more widespread, especially in the corporate sector. But companies increasingly want to encourage forms of collective practice that create value in more tangible ways: online ‘collectivities of work’ which involve people outside formal employment or organizational membership, in the interactive production of data-flows and activity-streams from which commercial value can be extracted. Let’s leave aside the intense debate about the status of this sort of labour or ‘playbor’ (),¹⁶ and focus on the underlying sociological question: what sort of collectivities does such ‘work’ construct on a massive scale? Take the case of so-called ‘brand volunteering’ (). It is performed by a complex set of figurations: the figurations of local brand communities (‘clubs’ with an interest for a certain brand) become linked to the figurations of online platforms and marketing campaigns to build up a certain collectivity of work. One study researched brand volunteering around a car company: anything like community only emerged at the local brand groups or during face-to-face events (). ‘Brand volunteering’ is marked by many tensions; mistrust by the supporters for the overall strategy of the company; feelings of being exploited. Nevertheless, the interest in the brand which motivates some members of brand communities to become ‘brand volunteers’ requires that the figuration built around the brand, and the broader context in which this ‘invitation to work’ occurs, is strong enough at least to encourage people to devote their scarce time to ‘collective’ projects oriented towards a reference-point that is abstract, a representation of a value.

For other ‘collectivities of work’ not even this is the case. They are built around online platforms to which individuals contribute ‘work’ as a collectivity of ‘online workers’ with little or no possibility of direct interaction with each other (they are not in this sense a ‘community’): for example, platforms which integrate individual contributions (rating platforms for people, restaurants, locations etc.) or platforms that link services that individuals offer (accommodation, rides etc.). In both cases these platforms are typically marketed as ‘cost saving’ and ‘democratizing’, but they are just as plausibly seen as forms of unpaid exploitation. The basic idea of rating platforms is that a company offers an online infrastructure to evaluate certain services. The unpaid work individuals do is to contribute their assessment to a public forum, with the assumed bonus that other users are paying attention to reviewers’ opinions. Such platforms then function as ‘a reflexive feedback loop’ (), both reflecting popular sentiments about specific services and helping to form those sentiments. From this perspective the actors and roles in this figuration are clearly defined by the supposed ‘benefit’ for all involved. The providers of this platform financially benefit by selling the information produced by this platform: the review-writers supposedly benefit by having the opportunity to express their opinion, and the readers, presumably, think they benefit by getting the latest information.

There is no communitization in this collectivity of (unpaid) work, whose outputs are sold by the platform operators. On the contrary, it can be argued that such collectivities of work have the side-effect of undermining community. See for example Zukin, Lindeman and Hurson’s () research on a rating platform for restaurants and clubbing locations that undermined existing community structures through the way it reinforced symbolic and other forms of privilege embedded in the practices of its ‘workers’. Far from being communitization, the side effects of this ‘collectivity of work’ may work to undermine actual processes of community building elsewhere. Similar dynamics operate in platforms that link together services that individuals offer (accommodation, car rides etc.). Such brokerage platforms are part of a ‘sharing economy’ (). The model of such platforms is to offer a convenient interface that matches supply and demand in trusted ways ().¹⁷ Such platforms have had far-reaching success in such varied areas as accommodation, car rides or jobs ().¹⁸ We can understand the collectivities structured around these platforms as ‘collectivities of work’ insofar as their members offer their hospitality, transport service or data labour for financial gain. But again such figurations are structured primarily around corporate platforms in order to make profit, with disturbing side-effects: for example, patterns of race segregation on accommodation platforms, including differential rates of remuneration driven by the automated collection of ethnicity data generated from the pictures users post of themselves (). We do not deny that, in some civil society platforms based on ‘sharing economy’ – including those close to social movements such as OpenStreetMaps () – the work is much less individualized and based on social movements and local groups with specific interests in such alternative platforms. Our point however concerns the general trend, which is dominated by the commercially successful platforms that follow a different kind of dynamic.

Other platforms produce a ‘crowd’ of low-skilled digital labourers who perform vast numbers of minor data tasks and are paid small amounts of money for this ‘microwork’ (). Such platforms organize digital workers to fit the needs of datafication. Service requesters post a ‘task’ that then can be performed by the ‘microworkers’, in accordance with the requirements of the programming or digital industries. On this model ‘humans (are) made into modular, protocol-defined computational services’ (), working on a large scale and in a fully monitorable way. Here, we start to see how the vast data infrastructure on which deep mediatization is based generates its own costs in terms of hidden forms of exploitative labour. Meanwhile, those forms of labour are themselves only possible on the basis of new types of figurations: collectivities without community. Through these forms of labour, individuals literally become part of the datafication process: ‘humans-as-a-service’ ().

Collectivities Through Numbers

There is a second way in which today’s data infrastructure generates new types of collectivity: ‘numeric inclusion’ (). There is nothing new about quantifying media audiences () or national populations (). But with datafication, precise measurement at one data-point can be calibrated, almost in real time, to information stored (or simultaneously gathered) in myriad other databases to produce instant categorization and ‘appropriate’ action. Based on the digital traces we all leave online, groupings of people who share certain characteristics are continuously produced to support the advertising industry’s aim of reaching individuals with customised advertising (). This accumulation of data is subsequently communicated back to the individuals: by online shops that produce lists of the other goods people making the same purchase bought; by online radio stations that produce charts, access statistics and rankings with reference to certain tastes of their users; by news pages that offer further information based on the previous reading choices of other users etc. Processes of ‘numeric inclusion’ therefore work to construct collectivities that ‘would not be possible without the measurement and activity assessments delegated to algorithms and statistical programs’ (). What is constructed here is an accumulation of individuals who are treated as sharing characteristics within networked processes of ‘big data’ processing.

‘Numeric inclusion’, however, when embedded back into daily practice, stimulates the construction for instrumental purposes of a meaningful horizon – mainly of taste, interest and orientation¹⁹ – within which individuals may come to position themselves as members of a certain collectivity:²⁰ those categorized as liking the same band, or the same type of book, or any other categorized feature. But such collectivities based on ‘numeric inclusion’ are generated without (or with very limited) communitization: indeed, there is no reference-point whatsoever for any affective relation to the collectivity’s other members, who remain hidden from each other, except via the assumption that they too were categorized the same way. Nevertheless, these collectivities are potentially influential: they represent large numbers of persons who have been constructed for important purposes as part of the same category, and who may orient their actions accordingly. To that extent, these collectivities too are an incipient contribution to an emerging social order.

Remaking Our Collectivities

At this stage it is an open question how much such collectivities formed by ‘numeric inclusion’ – that is, by datafication – might have in common with the media-based and mediatized local collectivities discussed at the beginning of this chapter. They would seem to be at opposite ends of the expanded spectrum of mediated collectivity in an age of deep mediatization. Are there practices emerging that might link up these different types of collectivity, and so transform the nature of communitization still further?

One important such practice is the embedding into everyday life of robotics, or rather ‘social robotics’: the ‘placement of robots in human social spaces’ (). At present, the most widespread forms of these ‘social robots’ are not physical artefacts but ‘assistant apps’ that can be associated with our smartphones (). Such apps are presented to us as quasi-human interaction partners with whom (through voice-recognition software) we can communicate using our natural voice. In reality, however, such ‘assistants’ are interfaces to large computer networks that process the questions we might ask and give ‘answers’ by reference to their comparability to an available dataset of questions and answers, combined with the data we have inputted. Through such a relationship we become a living, interactive member of a collectivity without community.

We can easily imagine face-to-face communities of people in such ‘relationships’ who come together to compare their experience of them. This may be only one of many entry-points for social robotics into the domain of social interaction: there has been much discussion of care-giving robots, or the robot puppets in dementia care,²¹ but just as relevant are ‘smart’ living environments and ‘smart’ self-driving cars and trains. Already, key aspects of smartphones and smart watches train us to communicate with systems of this sort.

There is of course no simple global story of how such new technologically mediated ways of constructing collectivity are everywhere transforming ‘the world’, ‘our world’ or ‘anyone’s world’: such stories remain rhetorics which mask hugely uneven processes that, in turn, are deeply implicated in underlying inequalities of socio-economic resource, which indeed they help deepen.²² Such rhetorics, even if already normalized in pioneer communities and the like, rely on very particular notions of ‘centre’ and ‘periphery’ that reduce most sites of everyday life across the globe to ‘sites of replication of a future invented prior and elsewhere’, as a recent analysis of IT practices in Peru notes ().

Our aim in registering some of the practices entangled with such rhetorics has absolutely not been to reproduce those rhetorics. But it would be equally dangerous not to acknowledge such pressures towards transformation, since they amount to nothing less than an attempt to transform our models of the social, a deeply motivated adjustment in the very basis of collectivity. As Sherry Turkle () puts it: ‘even before we make the robots, we remake ourselves as people ready to be their companions’. If this is so, we have no choice but, in important ways, to remake our collectivities and potentially our socialities too. But what are the wider costs? What are the consequences for the possibility of social order? These are the questions to which we turn in the book’s last two chapters.

Historical Parallels and Differences

To help us imagine the possible scale of changes in politics and government now under way, it is worth dwelling on a historical example when a shift in mediatization was associated with a change in the nature of government. Most accounts of early modernity focus on the invention from the fifteenth century onwards of printing in Europe as decisive for the later emergence of the modern nation-state. But more decisive, some historians argue, and more far-reaching in its implications for how modern government would come to be conducted, was the shift to rule through written document that M.T. Clanchy () dates much earlier, that is, to the eleventh to thirteenth centuries. The changing modalities of government were a crucial driver in cultural change: ‘lay literacy grew out of bureaucracy, rather than from any abstract desire for education or literature’ (). The increasing circulation of written documents for everyday purposes – from maps to legal records to royal decrees – and the skills associated with their production and interpretation, transformed how a ruler could interact with his territory. By the time of England’s King Edward I, government, in theory at least, ‘had access to every place of habitation, however small, and every man, however lowly his status’ (), not through any ease of personal physical access (roads remained very rudimentary), but through the coordinated power of a network of practices based on written documents. The state’s leading role in this shift parallels Elias’ account of the state’s leading role in the ‘structure of civilized behaviour’ and ‘manners’ () in the fifteenth and sixteenth centuries in Europe. The increasing shift from modes of government based on norms of personal (face-to-face) command to modes of government based on the deferred force of written description and instruction had other implications too: the growth of permanently available written records as reference-points for rule (); the need to find ways of ordering the sudden mass of important records ().⁴

The broad parallels with the profound transformation of government and politics through the explosion of digital records from the 1990s onwards () are striking. So too is the emergence of extreme inequalities of resource associated with both historic and contemporary transformations: just as today the power to build, design and manage digital infrastructures is concentrated in very few rich institutions (Google, Facebook, Amazon: see Chapter 4), so in the early medieval period ‘only the king possessed permanently organized writing facilities’ (). The potential for broader shifts in the organization of power and order from shifts in the degree of mediatization were present in the early medieval period, just as they are today. Those shifts operate within nations and between nations, with terms such as the ‘digital underclass’ () having an ambiguous status as a national and transnational phenomenon.

If by contrast we turn back to nineteenth-century America, we can learn also from James Beniger’s account of the emergence of modern communications – the telegraph, newspaper, postal services, telephone – in response to the ‘crisis of control’ created by the mismatch between the communication ‘demands’ of an expanding social and economic domain and the ‘supply’ of a rudimentary communications infrastructure. The transformation in modes of communication that emerged across all areas of business and government in the second half of the nineteenth century in expanding capitalist economies was not a matter of inventing a single medium and embedding it in daily life, but involved shifts in the interdependence of all aspects of organizational life, what Beniger calls ‘the progressive layering of control levels’ ().

How much then can we learn from these earlier precedents for the transformation in social order now under way through the wave of digitalization? One key difference between then and now is the new salience, even dominance, of private corporations, a theme that so far we have approached more generally through our discussion of the institutionalization of selves, collectivities and the frameworks in which organizations operate. All the processes we are discussing here have their basis in the institutionally based infrastructure of connection we call ‘the internet’. Certainly, the state in various countries was enormously important in the early investments and planning of the internet as a connective infrastructure (), a contribution that in fact goes back deep into the history of the computer and the nineteenth-century calculative technologies that preceded it (). But today, while the state is a key beneficiary of new forms of digital rule, it is not the main driver of change. As Louise Amoore argues in her history of the new data-based systems of surveillance and tracking movements across borders:

It is not the advent of a war on terror that ushered in a raft of novel and unprecedented security technologies. In fact, quite the reverse is true: the events of 9/11, and the exceptional measures that were so quickly declared actually cleared space for the sovereign enrolment of mundane, ordinary technologies that had been used in the gathering of everyday transactions of bread and sausages for some eight years. ()

Amoore details the role of executives from IBM and UK food and clothing retailer Marks and Spencer in developing ‘association rules’ that would predict consumer behaviour better from the textual traces available to new data mining techniques (). The drive of marketers to track consumers more continuously and predictively (from which states in turn benefited) was part of the expansion of opportunities for corporate organizations that derived from the new digital infrastructure. Other pressures towards data mining, however, derived from the increasing problem for advertisers that flowed from the explosion of digital media content in circulation in the late 1990s, which made it increasingly difficult to claim that any one expensive piece of discrete advertising reached its audience more effectively than another (). Building on this, powerful media players, such as Facebook, are ‘reworking the fabric of the web’ to make it a space better integrated to their economic goals ().

From many directions then, the organizational implications of digitalization are having profound consequences for order and rule. Also important, from the point of view of balances of power, is the shift in control over the computing infrastructure on which the digital world’s communications infrastructure relies. Today’s economy depends on vast data-processing capacity that has become, in most respects, too expensive for individual corporations or national governments to own internally. The rise of ‘the cloud’ – that is, ‘the storage, processing and distribution of data, applications and services for individuals and organizations’ () – represents a massive shift of control away from the nation-state, which until a decade ago had broad control of its information resources (), and towards the small number of dominant corporations (Apple, Google, Microsoft) which manage the remote computer servers on which ‘the cloud’ exists. We are not arguing here that the result is necessarily more order, let alone an order that is better attuned to key normative values. Although some classic analyses see the growth of information infrastructure as crucial to how effective social order was achieved in different phases of history (), there are counter-examples, for example the ambiguous status of bureaucracy in the large-scale violence of the twentieth century. We will return to the normative questions in Chapter 11.

A New Type of Politics?

Alerted then to the possibility that the changes in the communications infrastructure constituted by the waves of digitalization (and datafication) may themselves constitute a significant shift in the nature of social order, how can we relate this to the stuff of everyday politics and government?

Apparent changes in the nature of politics derived from digital media have attracted most attention. There is no doubt that media platforms for social networking have made short-term political mobilization easier (see Chapter 9), while also intensifying the dynamics of political scandal (). But this is very different from concluding that the nature and balance of politics overall has been changed by these new communication tools.

There are at least four strong reasons for caution against rushing to such a conclusion. First, the aspects of politics that would appear to have been transformed by digital communications are principally the ‘negative’ ones, what Pierre Rosanvallon () calls ‘counter-democracy’, leaving the implications for other aspects of politics, for example, the long-term building of parties for political transformation, less clear or perhaps even negative (). Second, the large-scale narratives told of how politics has changed through digital media tend to be universal stories built on generalizations from a limited number of cases, most notable the USA and the UK. More careful comparative analyses of digital politics paint a more mixed picture, with very different implications in different political systems (). Third, once we begin to grasp the complexity of how the wave of digitalization has affected organizations, we cannot underestimate the complexity of how it has affected political organizations such as parties engaged in multi-level conflicts. The combination of increased production and archiving of politically relevant information and more distributed access within and beyond organizational boundaries to that same information probably reduces the stability on average of political institutions (). Meanwhile, extended networks of communication have enabled organizations, such as political parties working in elections, to join up multiple political actors in what Rasmus Kleis Nielsen calls ‘campaign assemblages’ for the purposes of organizing face-to-face communication with potential voters (). Not only does this reverse the prediction of 1990s campaigners that ‘personalized’ communication was redundant in modern digital campaigning, but it alerts us to the effectiveness, in the pressured context of elections, of figurations of ‘fundamentally and never fully integrated mutually dependent actors who then go their separate ways once the election is over’ (). In such contexts, once again, the balance between political power and commercial power may be surprising: ‘your local supermarket is very likely to be more organizationally and technologically sophisticated than your local congressional campaign’, notes Nielsen ().

A final reason for caution, moving in the other direction, is that the general proliferation of politically relevant information (just as with consumer-relevant information) makes it all the more difficult to predict the consequences of political communication by particular organizations. Not only is cultural and historical context important, but so too (and just as culturally variable, potentially) are citizens’ habits of selection from the vast information universe available to them to motivate their political choices (). Many fear that information selection regularly locks people in ‘bubbles’ that are reinforced by the operations of search engines in presenting information to web searchers that fits better with the searchers’ past patterns of information use (). Against this background, we need to know more about the figurations of solidarity and expression in which citizens of particular types at particular locations are, or are not, involved. An excessive emphasis on networked politics at the expense of more detailed research on the social figurations in which network interactions actually matter has so far held back research in this area.⁵

A Different Type of State?

Rather than expect as yet to find clear patterns in how, under conditions of deep mediatization, particular political conflicts work themselves out, it may be more promising to examine the changing conditions for general projects of political order, that is, for government. As we saw already, governments, the embodiment of the nation-state, are becoming deeply reliant on the private corporate interests that sustain today’s data infrastructure. That granted, what are the implications of deep mediatization for rule?

If we think of governments as very large figurations of figurations, then the implications of deep mediatization for organizations apply even more strongly to governments: there are new opportunities for coordination on an extended scale and with increased precision over time, facilitating government by delegation to a vast range of authorities. Domains of internal networked communication are fundamental to the expanding work of government. But government, of course, does not work in a vacuum. The requirement of government, particularly in democratic states, to retain a minimal, even high, level of legitimacy before its citizens necessarily exposes governments to the accelerated communication flows of the general media environment. Over a decade ago, German political scientist Thomas Meyer () pessimistically argued that the foreshortened time-scales of media narratives in the 24/7 digital news environment accelerated not only the time-scale of government reaction to news events, but also the cycle of policy generation, with corrosive impacts on the quality of policy itself. There are certainly accounts from former senior governments that would support that view (), and the beginnings of comparative empirical work that might give some support to these fears (). But more extensive empirical work is needed.

Meanwhile, far from the locations where policy is developed, citizens go on with their lives, but in a changed relationship to what they still imagine as ‘the state’, mediated by an increasingly vast data infrastructure. Two authoritative commentators, Engin Isin and Evelyn Ruppert, note that ‘although the internet may not have changed politics radically in the fifteen years [between 1998 and 2013], it has radically changed the meaning and function of being citizens with the rise of both corporate and state surveillance’ (). It is not just a matter of performing a citizen identity in a new range of situations: it is, more fundamentally, a matter of increasingly integrating the rhythms and patterns of individual practice into the demands of the new systems of data collection on which the state increasingly relies. We saw in Chapter 7 the significance for selves of the rise of deep forms of categorization through data processes, in Chapter 8 for collectivities and their imaginary of political communities. But what exactly are the implications for government, and the possibility of social order through governing? We are, once more, too early in this shift to offer a definitive judgement, but some directions of change are becoming clear.

The importance of managing media for government’s stability has been clear for decades (), and has always revolved around the management presentation of external realities, such as crises, scandals and natural disasters. But as the project of government shifts in part to managing the vast data infrastructure on which its targeted actions-at-a-distance now depend, government-citizen relations themselves become increasingly mediated by this infrastructure and its bureaucratic interfaces. As Ruppert puts it:

What people do in relation to government (transactions) becomes more central rather than what they say they do and who they say they are (subjective identifications) [. . .] through all of this doing, subjects are less able to challenge, avoid or mediate their data double. ()

We discussed the concept of ‘data double’ in Chapter 7 in relation to the self, but we must return to it here since it is fundamental to understanding the type of governmental ‘order’ that is emerging under conditions of deep mediatization. Relations between citizens and governments have never been free from possibilities of alienation. But when governments’ actions, whatever their democratic intent, become routinely dependent on processes of automated categorization, a dislocation is threatened between citizens’ experience and the data trajectory on the basis of which they are judged. Schutz and Luckmann’s phenomenology grounded the ‘natural attitude of daily life’ in the individual’s ‘stock of previous experience’ () yet there was more than enough grim experience in the mid twentieth century to show that citizens could not always expect their governments to avoid actions that clashed violently with their stock of experience. But datafication is disturbing in a more banal way, because it is a mode of government that is never likely to be responsive to ‘what [citizens] say they do and who [citizens] say they are’.⁶

What consequences can we expect, over the longer term, for the core constituents of order and legitimate government: trust, legitimacy, the credibility of government action, citizens’ sense of their own efficacy as political actors? As yet, we do not know. But two large-scale and disturbing trends need at least to be registered. First, there is the association of the new datafied social infrastructure with the entrenchment of social economic and ethnic inequality (). Second, as already noted, there is the increasing importance of force, rather than consensus-based norms, in the constitution of social order. Norms in the area of data collection are, as we have seen, changing: dataism is precisely an attempt to inculcate new norms of data generation, that naturalize a certain submission to new relations between corporations and consumers, and citizens and state, based on continuous surveillance and the citizen’s active role in the generation of data. Yet the very infrastructure on which our networked digital communications relies is, in an important sense, ‘authoritarian’ (), based on ‘compliant submission to authority’, not consent. Norms in the absence of possible meaningful consent and negotiation become very close to force, with potentially dangerous long-term implications for wider institutional legitimacy.

That is our concern: that, under conditions of deep mediatization, an ever more complex infrastructure of interdependent communication installs a datafied social order which relies more on infrastructural force (or near-force) than on the openly contestable legitimacy of norms. We need in the next and concluding chapter to consider more fully the normative implications of such large-scale shifts in the construction of the social world and of social order.